{
  "schema_version": "1.0",
  "title": "Talent247 Jobs — Public Mandate Catalog",
  "description": "Live specialist roles across Artificial Intelligence, Machine Learning, AI Infrastructure, AI Research, MLOps, and Robotics.",
  "site_url": "https://jobs.talent247.co",
  "feed_url": "https://jobs.talent247.co/jobs.json",
  "llms_txt_url": "https://jobs.talent247.co/llms.txt",
  "llms_full_txt_url": "https://jobs.talent247.co/llms-full.txt",
  "agent_card_url": "https://jobs.talent247.co/.well-known/agent-card.json",
  "talent_pool_url": "https://jobs.talent247.co/talent-pool",
  "talent_pool_markdown_url": "https://jobs.talent247.co/talent-pool.md",
  "talent_pool_submit_url": "https://jobs.talent247.co/api/public/talent-pool",
  "content_signal": "search=yes, ai-input=yes, ai-train=no",
  "generated_at": "2026-11-16T10:46:20.339Z",
  "total_open_roles": 100,
  "facets": {
    "workplaces": [
      {
        "name": "Hybrid",
        "count": 78
      },
      {
        "name": "On-site",
        "count": 20
      },
      {
        "name": "Remote",
        "count": 2
      }
    ],
    "departments": [
      {
        "name": "Machine Learning",
        "count": 23
      },
      {
        "name": "Applied AI",
        "count": 21
      },
      {
        "name": "AI Infrastructure",
        "count": 18
      },
      {
        "name": "AI Research",
        "count": 15
      },
      {
        "name": "MLOps & Platform",
        "count": 14
      },
      {
        "name": "Robotics & Embodied AI",
        "count": 9
      }
    ],
    "employmentTypes": [
      {
        "name": "Full-time",
        "count": 97
      },
      {
        "name": "Contract",
        "count": 3
      }
    ],
    "locations": [
      {
        "name": "San Francisco, CA",
        "count": 24
      },
      {
        "name": "London, UK",
        "count": 17
      },
      {
        "name": "New York, NY",
        "count": 10
      },
      {
        "name": "Paris, France",
        "count": 8
      },
      {
        "name": "Seattle, WA",
        "count": 5
      },
      {
        "name": "Cambridge, UK",
        "count": 4
      },
      {
        "name": "Redwood City, CA",
        "count": 3
      },
      {
        "name": "Toronto, ON, Canada",
        "count": 3
      },
      {
        "name": "Austin, TX",
        "count": 2
      },
      {
        "name": "Berlin, Germany",
        "count": 2
      },
      {
        "name": "Montreal, QC, Canada",
        "count": 2
      },
      {
        "name": "Oxford, UK",
        "count": 2
      },
      {
        "name": "Palo Alto, CA",
        "count": 2
      },
      {
        "name": "Stockholm, Sweden",
        "count": 2
      },
      {
        "name": "Zurich, Switzerland",
        "count": 2
      },
      {
        "name": "Amsterdam, Netherlands",
        "count": 1
      },
      {
        "name": "Berlin / Cologne, Germany",
        "count": 1
      },
      {
        "name": "Boston, MA",
        "count": 1
      },
      {
        "name": "Bristol, UK",
        "count": 1
      },
      {
        "name": "Cambridge, MA",
        "count": 1
      },
      {
        "name": "Dublin, Ireland",
        "count": 1
      },
      {
        "name": "Edinburgh / Remote (UK)",
        "count": 1
      },
      {
        "name": "Freiburg / Berlin, Germany",
        "count": 1
      },
      {
        "name": "Munich, Germany",
        "count": 1
      },
      {
        "name": "Remote (US)",
        "count": 1
      },
      {
        "name": "Sunnyvale, CA",
        "count": 1
      },
      {
        "name": "Warsaw, Poland",
        "count": 1
      }
    ],
    "salaryBounds": {
      "min": 80000,
      "max": 395000
    },
    "remoteCount": 2,
    "featuredCount": 0
  },
  "jobs": [
    {
      "id": "da3a54c8-28d8-4152-ac50-beb80cae9e4c",
      "slug": "senior-distributed-data-vector-indexing-engineer",
      "title": "Senior Distributed Data & Vector Indexing Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 215000,
        "max": 305000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$215k – $305k",
        "subtitle": "annual package"
      },
      "summary": "Senior Distributed Data & Vector Indexing Engineer (MLOps & Platform) — New York, NY · Hybrid · $215k – $305k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier real-time AI search and knowledge discovery platform headquartered in San Francisco with a fast-growing systems engineering hub in New York City. Answering breaking-news, market-moving financial, and technical queries accurately requires ingesting, rendering, deduplicating, embedding, and indexing millions of newly published or updated web pages every hour—with end-to-end crawl-to-searchable freshness measured in seconds rather than days.\n\nThey have engaged Talent247 to appoint a Senior Distributed Data & Vector Indexing Engineer in New York, NY. Joining their Search Infrastructure & Freshness Indexing team, the successful candidate will architect petabyte-scale streaming pipelines and hybrid lexical-vector indexers in Rust, Go, and C++.\n\nKEY RESPONSIBILITIES\n• Architect ultra-high-throughput streaming ingestion and content-extraction pipelines in Rust and Go processing hundreds of thousands of web documents, feeds, and structured datasets per second.\n• Build real-time incremental vector and inverted-index storage tiers supporting sub-minute document visibility, instant tombstone deletions, and lock-free concurrent reads at tens of thousands of QPS.\n• Optimize batch-and-streaming GPU embedding inference pipelines (dynamic micro-batching, INT8/binary vector quantization, delta-chunk re-embedding) minimizing GPU cost across hundreds of billions of passages.\n• Design petabyte-scale content deduplication (SimHash/MinHash/URL canonicalization), spam/SEO-farm pre-filtering, and crawl-priority frontier schedulers reactive to live query spikes.\n• Engineer tiered RAM / PCIe Gen5 NVMe / object-store index layouts (DiskANN / HNSW / custom LSM-inverted structures) balancing sub-30ms tail latency with aggressive infrastructure efficiency.\n• Instrument end-to-end freshness SLAs, partition rebalancing automation, and multi-region active-active index replication.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust, Go, C++20, Kafka / Redpanda, Ray Data, Apache Arrow / Parquet / Lance, custom HNSW/DiskANN/BM25 storage engines, RocksDB, ClickHouse, Kubernetes, bare-metal NVMe & GPU clusters. Workplace: Hybrid in New York, NY. Compensation: $215,000 – $305,000 base salary plus high-upside unicorn equity and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of distributed systems, search-index infrastructure, or high-throughput streaming data platform engineering experience in Rust, C++, or Go.\n• Deep systems expertise in storage engines, LSM-trees, inverted indices, or approximate nearest-neighbor vector indices (HNSW, IVF-PQ, Vamana/DiskANN) operating across terabytes of NVMe/RAM.\n• Proven track record building low-latency streaming pipelines (Kafka/Redpanda/Flink/custom async Rust) sustaining hundreds of megabytes to gigabytes per second of sustained throughput.\n• Strong Linux performance engineering skills across memory-mapped IO (`mmap`), `io_uring`, SIMD acceleration, network serialization, and tail-latency debugging.\n• High ownership mindset thriving in lean systems squads operating planet-scale data infrastructure.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience building web crawlers, real-time news/financial search indices, or billion-scale vector databases (Qdrant, Milvus, Vespa, Lucene/Tantivy internals).\n• Experience orchestrating high-throughput GPU embedding/classification fleets alongside CPU streaming pipelines.\n• Open-source contributions to Rust search/storage ecosystems (`tantivy`, `lancedb`, `qdrant`, `rocksdb`, `arrow-rs`).",
      "updated_at": "2026-11-15T17:43:25.785Z",
      "created_at": "2026-11-15T16:51:11.969Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-distributed-data-vector-indexing-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-distributed-data-vector-indexing-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-distributed-data-vector-indexing-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-distributed-data-vector-indexing-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "21ab6523-d1db-4e52-8d12-7d84a4efa9ee",
      "slug": "senior-post-training-ml-engineer-grounded",
      "title": "Senior Post-Training ML Engineer, Grounded Reasoning",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 240000,
        "max": 340000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$240k – $340k",
        "subtitle": "annual package"
      },
      "summary": "Senior Post-Training ML Engineer, Grounded Reasoning (Machine Learning) — San Francisco, CA · On-site · $240k – $340k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a category-defining San Francisco AI search, answer, and autonomous research company serving tens of millions of users and thousands of API developers worldwide. Their in-house Post-Training & Reasoning Models group trains custom low-latency grounded models and deep multi-step research agents that interleave chain-of-thought reasoning with live web search, code execution, and strict claim-by-claim citation verification.\n\nThey have partnered with Talent247 to hire a Senior Post-Training ML Engineer for Grounded Reasoning onsite in San Francisco, CA. The appointed engineer will design online reinforcement learning (GRPO/PPO) environments and faithfulness reward pipelines that teach models when to search, how to reconcile conflicting live sources, and how to synthesize zero-hallucination answers at blazing speed.\n\nKEY RESPONSIBILITIES\n• Design and scale end-to-end post-training pipelines—combining SFT, DPO, and online multi-turn RL (GRPO/PPO/RLOO)—over dense and Mixture-of-Experts open-weight and custom pretrained backbones.\n• Build live, cached-deterministic web-search and code-interpreter RL environments where models learn adaptive query decomposition, source cross-checking, date-sensitive reasoning, and efficient search-budget stopping.\n• Train high-precision generative reward models and claim-entailment verifiers that penalize unsupported extrapolation, stale facts, sycophancy, and prompt-injected web spam.\n• Distill multi-minute deep-research reasoning trajectories into compact, ultra-fast production models delivering sub-second time-to-first-token at massive consumer QPS.\n• Collaborate closely with neural retrieval engineers to co-train search query planners and answer synthesizers end-to-end.\n• Maintain rigorous live-web, multilingual, financial, coding, and adversarial factuality leaderboards gating every production model rollout.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Verl / OpenRLHF / Megatron-Core / FSDP2, SGLang/vLLM/TensorRT-LLM, Ray, ClickHouse, multi-thousand H100/B200 GPU clusters. Workplace: Onsite in San Francisco, CA. Compensation: $240,000 – $340,000 base salary plus substantial unicorn equity, daily catered meals, and full healthcare & 401(k) benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of hands-on ML engineering or research engineering experience in LLM post-training, RLHF/RLVR, or tool-augmented reasoning models.\n• Proven production experience scaling distributed SFT, DPO, and online RL (GRPO/PPO) runs across 30B–400B+ dense and MoE transformer architectures.\n• Deep familiarity with multi-turn tool-use RL mechanics—asynchronous rollout generation with vLLM/SGLang, environment state caching, off-policy importance sampling, and reward-hacking prevention.\n• Strong instinct for factuality evaluation, citation entailment modeling, and diagnosing subtle hallucination modes under conflicting retrieved evidence.\n• High-velocity execution mindset comfortable shipping model iterations from research ablation to millions-of-users production within weeks.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience training DeepResearch-style browsing agents, search-integrated RL models (Search-R1-class paradigms), or code-execution reasoning models.\n• Experience with on-policy distillation or speculative decoding draft-model training for latency-critical consumer serving.\n• Publications at NeurIPS, ICML, ICLR, ACL, or EMNLP in post-training, RLHF, or grounded generation.",
      "updated_at": "2026-11-15T17:43:24.596Z",
      "created_at": "2026-11-15T16:51:08.791Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-post-training-ml-engineer-grounded",
        "markdown": "https://jobs.talent247.co/jobs/senior-post-training-ml-engineer-grounded.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-post-training-ml-engineer-grounded/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-post-training-ml-engineer-grounded/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "e9f01dad-dfa5-4a8c-9ebe-a28ab3290fba",
      "slug": "distributed-storage-engineer-model-weight",
      "title": "Distributed Storage Engineer, Model Weight Streaming",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "New York, NY",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 225000,
        "max": 320000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $320k",
        "subtitle": "annual package"
      },
      "summary": "Distributed Storage Engineer, Model Weight Streaming (AI Infrastructure) — New York, NY · On-site · $225k – $320k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier New York City serverless GPU and AI compute platform powering instant-scaling inference, batch multimedia processing, and massive reinforcement learning sandboxes. When a customer scales a 400-gigabyte Mixture-of-Experts model from zero to 500 GPU replicas in three seconds, conventional container registries and POSIX network filesystems collapse—making their custom Rust distributed storage engine a foundational competitive moat.\n\nThey have partnered with Talent247 to hire a Distributed Storage Engineer for Model Weight Streaming onsite in New York, NY. The successful candidate will co-architect their content-addressable filesystem, tiered NVMe-oF caches, and zero-copy RDMA/PCIe weight-streaming protocols saturating hundreds of gigabits per second per GPU node.\n\nKEY RESPONSIBILITIES\n• Design and scale a custom Rust distributed, content-addressable storage and volume engine serving petabytes of model weights, container layers, and training checkpoints across tens of thousands of concurrent serverless workers.\n• Engineer ultra-high-throughput tiered caching hierarchies spanning local PCIe Gen5 NVMe arrays, peer-to-peer cluster caches, erasure-coded regional storage nodes, and cloud object stores.\n• Implement zero-copy `io_uring`, direct IO (`O_DIRECT`), and GPUDirect Storage (`cuFile` / DMA) streaming paths loading `safetensors` shards from network/NVMe directly into GPU HBM at line rate.\n• Build transactional copy-on-write (CoW) volume snapshotting and instant branching primitives enabling million-container RL coding/data sandboxes to fork terabyte datasets in milliseconds.\n• Design consensus, metadata sharding, background garbage collection, and bit-rot scrubbing protocols upholding strict durability and consistency guarantees under node churn.\n• Benchmark and model tail-latency behavior under extreme thundering-herd fan-out workloads using deterministic simulation testing.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust, `io_uring`, FUSE / NFS / custom RPC protocols, NVMe / NVMe-oF, RDMA, erasure coding (Reed-Solomon), Raft/Paxos metadata stores, S3/GCS backends, Linux bare-metal & cloud GPU fleets. Workplace: Onsite in New York, NY. Compensation: $225,000 – $320,000 base salary plus substantial equity ownership, full healthcare, daily meals, and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of distributed storage, filesystem, database storage-engine, or high-performance networked IO engineering experience in Rust, C++, or C.\n• Deep understanding of Linux storage stacks (`io_uring`, NVMe block layer, page cache vs. `O_DIRECT`, FUSE, filesystem metadata layouts) and networked storage throughput tuning.\n• Strong distributed systems foundations across consistent hashing, replication, erasure coding, lease/lock management, and crash recovery.\n• Proven ability to reason quantitatively about disk IOPS, network bisection bandwidth, read/write amplification, and thundering-herd cache stampedes.\n• Passion for writing clean, deterministic, simulation-tested Rust systems code.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with GPUDirect Storage, RDMA/RoCEv2 data transfers, or P2P content-addressed distribution of massive ML artifacts (`safetensors`, GGUF, OCI layers).\n• Prior work on distributed filesystems or object/block stores (Ceph, Lustre, 3FS/Tectonic/JuiceFS-class architectures, FoundationDB, RocksDB).\n• Experience with deterministic simulation testing (FoundationDB / `turmoil`-style fault injection) in Rust.",
      "updated_at": "2026-11-15T17:43:23.856Z",
      "created_at": "2026-11-15T16:50:36.078Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/distributed-storage-engineer-model-weight",
        "markdown": "https://jobs.talent247.co/jobs/distributed-storage-engineer-model-weight.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/distributed-storage-engineer-model-weight/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/distributed-storage-engineer-model-weight/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "174c4fdb-559f-4211-a66e-1bbb44c57c2e",
      "slug": "senior-3d-perception-sensor-fusion-ml-engineer",
      "title": "Senior 3D Perception & Sensor Fusion ML Engineer",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "Oxford, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 85000,
        "max": 120000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£85k – £120k",
        "subtitle": "annual package"
      },
      "summary": "Senior 3D Perception & Sensor Fusion ML Engineer (Robotics & Embodied AI) — Oxford, UK · Hybrid · £85k – £120k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an Oxford-headquartered autonomous vehicle software leader commercializing universal, safety-certified autonomy across industrial logistics yards, airports, ports, mines, and commercial passenger shuttles. Unlike fair-weather highway demos, their fleets operate 24/7 in complex, dynamic industrial environments alongside heavy machinery and pedestrians under heavy rain, fog, dust, and night-time conditions.\n\nThey have partnered with Talent247 to hire a Senior 3D Perception & Sensor Fusion ML Engineer in Oxford. Joining their Core Perception squad, the appointed engineer will design, train, and deploy multi-modal 3D bird's-eye-view (BEV) detection, 4D occupancy, and multi-object tracking networks fusing camera, LiDAR, and imaging-radar streams onto onboard vehicle compute.\n\nKEY RESPONSIBILITIES\n• Architect and train spatiotemporal multi-sensor fusion models (BEV transformers, 4D semantic occupancy networks, point-cloud + image fusion) delivering centimetre-accurate 3D understanding in unstructured and adverse-weather environments.\n• Design self-supervised pre-training, auto-labeling teacher distillation, and active-learning loops over terabytes of multi-fleet driving logs to slash manual 3D cuboid annotation costs.\n• Optimize onboard model architectures via quantization-aware training (INT8/FP16), structured pruning, and TensorRT/C++ graph compilation to hit strict real-time latency and determinism budgets on vehicle GPUs.\n• Build granular safety-critical evaluation slices targeting occluded pedestrians, degraded sensor dropouts, calibration perturbations, and rare industrial obstacles.\n• Collaborate closely with sensor-calibration, motion-planning, and onboard systems engineers from lab hardware-in-the-loop rigs through live test-track and customer fleet rollouts.\n• Mentor fellow perception engineers and contribute to architectural roadmaps across next-generation sensor suites.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir perception stack runs on Python, PyTorch, C++17/20, CUDA/TensorRT, ROS2/custom deterministic middleware, and cloud GPU training clusters. Based hybrid in Oxford, our client offers £85,000–£120,000 GBP base salary plus equity options, generous UK pension, 28 days holiday + bank holidays, private healthcare, and UK Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of industry experience in 3D computer vision and deep learning perception for autonomous vehicles, mobile robotics, or embodied AI.\n• Deep hands-on proficiency in PyTorch for training 3D object detection, segmentation, occupancy, or multi-object tracking networks across LiDAR, camera, and/or radar modalities.\n• Solid C++ and CUDA/TensorRT skills for optimizing and integrating trained perception models into real-time onboard vehicle pipelines.\n• Strong geometric fundamentals in camera projection models, extrinsic/intrinsic sensor calibration, ego-motion compensation, and temporal fusion.\n• Proven ability to diagnose real-world fleet failure modes and translate them into measurable model and dataset improvements.\n\nPREFERRED / BONUS EXPERIENCE\n• MSc or PhD in Robotics, Computer Vision, or Machine Learning, or publications at CVPR, ICCV, ECCV, ICRA, or IROS.\n• Experience with foundation vision backbones, distillation from offboard 3D auto-labelers, or safety-critical ISO 26262 / SOTIF validation workflows.",
      "updated_at": "2026-11-15T17:42:57.583Z",
      "created_at": "2026-11-15T16:54:01.087Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "818aaac8-a1bf-4985-9edd-4753a310c8bf",
      "slug": "senior-real-time-voice-agent-streaming-systems",
      "title": "Senior Real-Time Voice Agent & Streaming Systems Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 105000,
        "max": 145000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£105k – £145k",
        "subtitle": "annual package"
      },
      "summary": "Senior Real-Time Voice Agent & Streaming Systems Engineer (Applied AI) — London, UK · Hybrid · £105k – £145k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally renowned generative audio and voice AI unicorn headquartered in London and New York, powering real-time conversational voice agents for technology platforms, healthcare providers, financial services, and interactive entertainment worldwide. In live duplex voice interaction, every 20 milliseconds of network jitter, turn-detection delay, or tool-execution lag separates a magical human-like conversation from an awkward robotic pause.\n\nThey have engaged Talent247 to hire two Senior Real-Time Voice Agent & Streaming Systems Engineers in London. Joining their Conversational Agents Infrastructure group, the appointed engineers will architect the globally distributed WebRTC/SIP media plane and stateful streaming orchestration engine coordinating sub-200ms speech recognition, LLM reasoning, MCP/webhook tool calls, and expressive neural audio synthesis across hundreds of thousands of concurrent calls.\n\nKEY RESPONSIBILITIES\n• Architect, scale, and operate globally distributed, edge-routed WebRTC and SIP telephony media servers in Rust, Go, and Python delivering ultra-low-jitter bidirectional audio streaming.\n• Engineer stateful real-time conversation loops handling semantic turn-end prediction, natural back-channeling, mid-sentence user barge-in cancellation, and dynamic background-noise suppression.\n• Build deterministic asynchronous tool-execution brokers (supporting webhooks, client-side SDK callbacks, and Model Context Protocol servers) that let agents fetch data or execute actions mid-call while gracefully filler-bridging or streaming interim speech.\n• Instrument millisecond-level end-to-end latency waterfalls (network RTT, VAD/turn-taking, ASR, LLM TTFT, TTS first-chunk) across global regions using OpenTelemetry and ClickHouse.\n• Design chaos-engineering, call-replay debugging, and high-concurrency load-testing harnesses ensuring 99.99% uptime under sudden enterprise call spikes.\n• Partner directly with audio ML researchers to co-design native speech-to-speech streaming protocols replacing cascaded text bottlenecks.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir systems stack runs on Rust, Go, Python, WebRTC, Pion/LiveKit/custom RTP media engines, SIP trunks, Kubernetes, ClickHouse, Redis, and GPU inference clusters. Working hybrid in central London, our client offers £105,000–£145,000 GBP base salary plus substantial unicorn equity, private healthcare, and UK visa/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of backend and real-time distributed systems engineering experience building latency-critical streaming media, telephony, gaming, or real-time AI platforms.\n• Strong systems programming skills in Rust or Go alongside Python, with deep mastery of asynchronous concurrency, lock-free audio ring buffers, and network tail-latency tuning.\n• Direct production experience with WebRTC (ICE/STUN/TURN, Opus/PCM codecs, jitter buffers), WebSockets, gRPC bidirectional streaming, and/or SIP telephony infrastructure.\n• Experience integrating real-time ML inference streams (streaming ASR, LLM token streams, chunked TTS, or neural VAD) under sub-300ms round-trip budgets.\n• High ownership mentality in a fast-shipping engineering culture where engineers own architecture from RFC through global production rollout.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience building conversational AI voice agents, CCaaS telephony bridges, or Model Context Protocol (MCP) tool runtimes.\n• Familiarity with embedded audio DSP (echo cancellation, noise suppression) or global Anycast/edge routing.",
      "updated_at": "2026-11-15T17:42:55.637Z",
      "created_at": "2026-11-15T16:53:38.445Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems",
        "markdown": "https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "61b70d66-76b2-40c4-aaf1-90e69e8a54cf",
      "slug": "senior-inference-engineer-real-time-diffusion",
      "title": "Senior Inference Engineer, Real-Time Diffusion Transformers",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 105000,
        "max": 150000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£105k – £150k",
        "subtitle": "annual package"
      },
      "summary": "Senior Inference Engineer, Real-Time Diffusion Transformers (AI Infrastructure) — London, UK · Hybrid · £105k – £150k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Britain's flagship $4B generative AI video unicorn headquartered in central London, serving more than 60,000 businesses worldwide with expressive neural video avatars and studio-grade generative video models. As their product evolves from asynchronous batch video rendering to live, conversational real-time video avatars that see, listen, react, and speak with sub-second latency, runtime inference efficiency on Diffusion Transformers (DiTs) has become a mission-critical engineering pillar.\n\nThey have retained Talent247 to appoint a Senior Inference Engineer, Real-Time Diffusion Transformers in London. Bridging their Generative Video Research group and Production Inference Infrastructure squad, the appointed engineer will take multi-billion-parameter causal audio-conditioned video diffusion and autoregressive models and make them stream infinitely long 1080p+ video sequences faster than real time on GPU clusters.\n\nKEY RESPONSIBILITIES\n• Profile, optimize, and deploy causal streaming Diffusion Transformer (DiT) and 3D spatiotemporal VAE architectures to achieve sustained >30 FPS real-time interactive avatar generation at <300ms time-to-first-frame.\n• Implement custom CUDA, Triton, and TensorRT optimizations—including FP8/INT8/NVFP4 quantization, KV-cache rolling windows for causal video attention, CUDA Graphs, and fused audio-conditioning cross-attention kernels.\n• Design multi-GPU sequence/context parallelism (Ulysses / Ring Attention) and disaggregated VAE decoding pipelines scaling smoothly across H100, H200, and B200 accelerator pods.\n• Partner directly with research scientists on few-step adversarial/consistency distillation and causal architectural modifications that unlock 5x–10x runtime speedups without lip-sync or identity degradation.\n• Integrate ultra-low-latency GPU video encoding (NVENC) directly with WebRTC media servers for zero-copy GPU-to-network frame streaming.\n• Build automated perceptual-regression and frame-pacing benchmark suites guarding every runtime optimization.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack features Python, C++20, PyTorch (`torch.compile`), CUDA/CUTLASS, Triton, TensorRT, NVENC/WebRTC, and Kubernetes H100/B200 GPU fleets. Offered hybrid in central London, compensation includes £105,000–£150,000 GBP base salary plus Series E unicorn equity, comprehensive private healthcare, pension, and UK Skilled Worker / Global Talent visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in GPU performance engineering, deep learning inference optimization, or high-performance multimedia/computer-vision systems.\n• Deep hands-on mastery of PyTorch internals, CUDA/Triton kernel development, TensorRT, and hardware-level profiling using NVIDIA Nsight Systems and Nsight Compute.\n• Direct practical experience optimizing transformer or diffusion architectures (DiTs, latent video diffusion, 3D VAEs, or LLMs) via FP8/INT8 quantization, operator fusion, and multi-GPU parallelism.\n• Strong systems understanding of GPU memory hierarchies, PCIe/NVLink bandwidth, asynchronous CUDA streams, and tail-latency elimination.\n• Collaborative mindset excited to co-design hardware-efficient architectures side-by-side with generative video researchers.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with real-time video streaming pipelines (NVENC/NVDEC, GStreamer, WebRTC) or causal streaming video diffusion (FIFO-Diffusion / CausVid-style rolling caches).\n• Contributions to open-source GPU inference frameworks (`diffusers`, `xDiT`, `TensorRT`, `vLLM`, `SGLang`, or `FlashAttention`).",
      "updated_at": "2026-11-15T17:42:54.883Z",
      "created_at": "2026-11-15T16:53:26.964Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-inference-engineer-real-time-diffusion",
        "markdown": "https://jobs.talent247.co/jobs/senior-inference-engineer-real-time-diffusion.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-inference-engineer-real-time-diffusion/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-inference-engineer-real-time-diffusion/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "e6f1b328-24f1-41e8-a685-88dd8084a471",
      "slug": "senior-systems-engineer-ultra-low-latency",
      "title": "Senior Systems Engineer, Ultra-Low-Latency Inference Cloud",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Austin, TX (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 185000,
        "max": 255000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$185k – $255k",
        "subtitle": "annual package"
      },
      "summary": "Senior Systems Engineer, Ultra-Low-Latency Inference Cloud (AI Infrastructure) — Austin, TX · Hybrid · $185k – $255k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a breakthrough custom AI silicon and inference-cloud provider operating massive dedicated AI supercomputing clusters across North America, with a flagship systems engineering hub in Austin, TX. Because their custom accelerators generate LLM tokens at 2,000+ tokens per second per stream—an order of magnitude faster than conventional GPU clouds—standard HTTP gateways and Python serving wrappers immediately become the bottleneck.\n\nThey have engaged Talent247 to hire a Senior Systems Engineer for their Ultra-Low-Latency Inference Cloud in Austin, TX. The appointed engineer will build ultra-lean Rust/C++/Go streaming data planes, stateful KV-cache-aware request routers, and multi-region failover fabrics capable of sustaining millions of tokens per second with single-digit millisecond gateway overhead.\n\nKEY RESPONSIBILITIES\n• Architect high-concurrency, zero-copy Rust and Go token-streaming gateways (HTTP/2, HTTP/3, gRPC, SSE, WebSockets) adding <2ms p99 overhead on 2,000+ tok/s reasoning and coding streams.\n• Design distributed prefix-cache-aware and session-affine request schedulers that maximize on-silicon KV-cache hit rates across multi-turn agentic coding and tool-calling workloads.\n• Build continuous-batching admission controllers, multi-tier token-bucket rate limiters, and fairness queues balancing interactive low-latency API traffic against massive asynchronous batch/RL rollout jobs.\n• Engineer zero-downtime weight-reload orchestration and fast cross-datacenter traffic draining over bare-metal Kubernetes clusters connected directly to custom AI accelerators.\n• Instrument nanosecond-to-millisecond distributed tracing (OpenTelemetry, eBPF) tracking time-to-first-token (TTFT), inter-token latency (ITL), and queue wait across every hop.\n• Partner closely with compiler and runtime engineers to expose structured outputs, speculative decoding, and reasoning-effort controls cleanly via industry-standard streaming chat-completion APIs.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust, Go, C++20, Tokio/io_uring, Envoy, gRPC, Kubernetes (bare-metal), Redis/Dragonfly, ClickHouse, Prometheus/eBPF, Linux high-speed networking. Workplace: Hybrid in Austin, TX. Compensation: $185,000 – $255,000 base salary plus meaningful equity, relocation assistance, and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems, networking, or cloud data-plane engineering experience building ultra-low-latency, high-throughput distributed services in Rust, C++, or Go.\n• Deep understanding of Linux networking internals, async event loops (Tokio/epoll/io_uring), HTTP/2 & gRPC framing, backpressure handling, and tail-latency elimination.\n• Production experience operating stateful or hardware-attached services on Kubernetes, custom schedulers, or Envoy/service-mesh control planes.\n• Familiarity with LLM inference serving mechanics—including continuous batching, prefill vs. decode disaggregation, KV-cache locality, and streaming tokenization.\n• Strong operational discipline around load testing, flame-graph profiling, and multi-region high availability.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct contributions to LLM serving routers or gateways (vLLM router, SGLang router, Dynamo, Envoy AI Gateway, Pingora).\n• Experience with RDMA/RoCEv2 networking, DPDK/XDP kernel-bypass, or custom hardware host-interface drivers.\n• Track record scaling developer-facing commercial API clouds under strict 99.99% uptime SLAs.",
      "updated_at": "2026-11-15T17:42:49.733Z",
      "created_at": "2026-11-15T16:52:14.819Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-systems-engineer-ultra-low-latency",
        "markdown": "https://jobs.talent247.co/jobs/senior-systems-engineer-ultra-low-latency.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-systems-engineer-ultra-low-latency/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-systems-engineer-ultra-low-latency/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "6499b45f-e623-40f5-958f-9651c7103085",
      "slug": "forward-deployed-ai-engineer-inference-fine-tuning",
      "title": "Forward-Deployed AI Engineer, Inference & Fine-Tuning",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 195000,
        "max": 275000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$195k – $275k",
        "subtitle": "annual package"
      },
      "summary": "Forward-Deployed AI Engineer, Inference & Fine-Tuning (Applied AI) — New York, NY · Hybrid · $195k – $275k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a high-growth AI inference and customization cloud headquartered on the San Francisco Peninsula with a flagship East Coast Applied Engineering hub in New York City. Fast-scaling AI startups and Fortune 500 engineering organizations rely on them to migrate massive token-volume workloads from slow, rate-limited proprietary APIs onto blazing-fast, custom post-trained open-weight models.\n\nThey have engaged Talent247 to recruit a Forward-Deployed AI Engineer specializing in Inference & Fine-Tuning in New York, NY. Embedded directly with top customer ML and platform engineering teams, the successful candidate will own end-to-end model distillation, reinforcement fine-tuning, and sub-second serving optimization across flagship production deployments.\n\nKEY RESPONSIBILITIES\n• Partner hands-on with customer Staff/Principal ML engineers to benchmark, distill, and fine-tune open-weight dense and MoE models (via SFT, DPO, and GRPO reinforcement fine-tuning) that surpass proprietary frontier baselines on their exact domain tasks.\n• Build golden evaluation harnesses, custom programmatic/rubric graders, and synthetic trajectory curation pipelines over real customer production logs (under strict VPC/zero-retention controls).\n• Co-optimize customer prompt topologies for prefix-cache hit rates, grammar-constrained JSON schemas, parallel tool-calling, and custom EAGLE/ngram speculative decoding configurations.\n• Diagnose end-to-end TTFT and tokens-per-second bottlenecks across multi-region dedicated GPU deployments, tuning tensor/expert parallelism and batching SLAs to traffic shapes.\n• Prototype compound multi-model architectures (fast semantic routers + specialized LoRA adapters + verifier models) cutting customer inference spend by 5x–10x while halving latency.\n• Channel field insights directly into core inference engine, fine-tuning API, and developer SDK features alongside Redwood City product engineers.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, TypeScript, PyTorch, SFT/DPO/GRPO fine-tuning APIs & custom recipes, industry-standard chat-completion & native gRPC streaming SDKs, vLLM/SGLang/TensorRT-class serving parameters, Kubernetes, AWS/GCP. Workplace: Hybrid in New York, NY. Compensation: $195,000 – $275,000 base salary plus substantial equity grants, performance incentives, and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of applied ML engineering, ML infrastructure, or forward-deployed AI engineering experience shipping high-throughput LLM systems into production.\n• Hands-on track record fine-tuning open-weight language or multimodal models (dense and sparse Mixture-of-Experts transformer architectures) using LoRA/full-weight SFT, DPO, or RLVR/GRPO.\n• Strong command of LLM inference mechanics—KV-cache sizing, prefix caching, quantization (FP8/FP4), tensor vs. pipeline parallelism, and speculative decoding trade-offs.\n• Production coding fluency in Python and TypeScript/Go, comfortable pairing directly inside customer repositories and evaluation harnesses.\n• High autonomy and crisp technical communication with VPs of Engineering and AI research leads.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience optimizing real-time voice-agent LLM backends, code-completion models, or high-QPS search/reranking pipelines.\n• Prior forward-deployed or applied solutions engineering tenure at an AI lab, inference cloud, or MLOps unicorn.\n• Contributions to open-source fine-tuning, evaluation, or structured-generation libraries.",
      "updated_at": "2026-11-15T17:42:45.774Z",
      "created_at": "2026-11-15T16:50:09.267Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning",
        "markdown": "https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "9031845b-6e8a-499e-ad70-eb8769400257",
      "slug": "principal-gpu-inference-kernel-engineer",
      "title": "Principal GPU Inference Kernel Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 300000,
        "max": 395000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$300k – $395k",
        "subtitle": "annual package"
      },
      "summary": "Principal GPU Inference Kernel Engineer (AI Infrastructure) — San Francisco, CA · On-site · $300k – $395k. THE CLIENT & MANDATE Our client is a premier…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier frontier AI research and deployment company in San Francisco serving trillions of tokens every day across interactive reasoning assistants, real-time voice streams, and developer APIs. At planetary inference scale, shaving microseconds off attention, MoE dispatch, and quantized GEMM kernels translates directly into tens of thousands of reclaimed GPUs and dramatically deeper test-time reasoning budgets for users worldwide.\n\nThey have partnered with Talent247 to appoint a Principal GPU Inference Kernel Engineer onsite in San Francisco, CA. Operating at the bare metal of NVIDIA Hopper, Blackwell, and next-generation accelerator architectures, the successful candidate will set the technical bar for custom CUDA/PTX kernels and disaggregated serving runtimes across flagship frontier models.\n\nKEY RESPONSIBILITIES\n• Design and hand-optimize state-of-the-art CUDA, PTX, CUTLASS, and CuTe C++ kernels for paged/sparse FlashAttention, grouped MoE GEMMs, fused activation-quantization, and speculative multi-token verification.\n• Push hardware utilization near theoretical tensor-core and HBM3e/HBM4 bandwidth ceilings using warp specialization, TMA asynchronous copies, persistent kernels, and NVFP4/FP8/MXFP microscaling formats.\n• Co-architect disaggregated prefill-and-decode serving topologies and ultra-low-latency NVLink / InfiniBand KV-cache transfer engines (custom NVSHMEM / GPUDirect RDMA collectives).\n• Partner directly with pre-training model architects to co-design hardware-friendly attention heads, MoE expert granularities, and KV-cache compression schemes 12 months ahead of launch.\n• Build deterministic cycle-level profiling harnesses, numerical accuracy verifiers, and automated autotuning pipelines across heterogeneous GPU SKUs.\n• Serve as an organization-wide technical authority mentoring senior kernel and compiler engineers across inference and RL rollout teams.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: C++20, CUDA, PTX/SASS inspection, CUTLASS 3.x / CuTe, Triton, NVSHMEM/NCCL, PyTorch custom ops, Nsight Compute/Systems, Hopper (H100/H200) & Blackwell (B200/GB200 NVL72) superclusters. Workplace: Onsite in San Francisco, CA. Compensation: $300,000 – $395,000 base salary plus exceptional Principal-level equity grants and comprehensive benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 8+ years of low-level GPU performance engineering, HPC kernel architecture, or deep learning inference runtime development in C++ and CUDA.\n• Mastery of modern NVIDIA GPU microarchitecture (Hopper WGMMA/TMA, Blackwell tcgen05/NVFP4, warp scheduling, register pressure, shared-memory swizzling, L2 persistence).\n• Track record authoring production kernels that outperform vendor library baselines (cuBLAS, cuDNN) on real transformer or MoE workloads.\n• Deep familiarity with multi-GPU communication primitives (NVLink, NVSwitch, NVSHMEM, GPUDirect RDMA) and CUDA Graph execution models.\n• Ability to bridge low-level SASS/PTX hardware realities with high-level model architecture trade-offs.\n\nPREFERRED / BONUS EXPERIENCE\n• Core authorship or major contributions to FlashAttention, FlashInfer, TensorRT-LLM, vLLM, SGLang, DeepEP, FasterTransformer, or CUTLASS.\n• Experience optimizing large-scale disaggregated prefill/decode serving or multi-node NVL72 rack-scale inference.\n• Compiler internals familiarity (Triton backend, MLIR, LLVM NVPTX) for custom code-generation DSLs.",
      "updated_at": "2026-11-15T17:42:40.931Z",
      "created_at": "2026-11-15T16:49:28.830Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/principal-gpu-inference-kernel-engineer",
        "markdown": "https://jobs.talent247.co/jobs/principal-gpu-inference-kernel-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/principal-gpu-inference-kernel-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/principal-gpu-inference-kernel-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "6beda249-225f-4551-ad9b-63e919de7f3d",
      "slug": "senior-ai-agent-autonomous-lab-orchestration",
      "title": "Senior AI Agent & Autonomous Lab Orchestration Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Boston, MA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 195000,
        "max": 265000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$195k – $265k",
        "subtitle": "annual package"
      },
      "summary": "Senior AI Agent & Autonomous Lab Orchestration Engineer (Applied AI) — Boston, MA · Hybrid · $195k – $265k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an ambitious autonomous-science AI company operating robotic discovery facilities across Greater Boston and Cambridge, MA. Their platform closes the full loop from AI hypothesis generation and literature synthesis through automated protocol compilation, robotic liquid-handler and analytical-instrument execution, and real-time multimodal readout ingestion.\n\nThey have engaged Talent247 to hire a Senior AI Agent & Autonomous Lab Orchestration Engineer in Boston, MA. Sitting at the critical boundary between frontier reasoning agents and million-dollar robotic laboratory workcells, the appointed engineer will build stateful orchestration harnesses and deterministic safety verifiers that translate abstract scientific hypotheses into collision-free, chemically safe physical execution.\n\nKEY RESPONSIBILITIES\n• Architect durable, stateful multi-agent orchestration pipelines that compile high-level experimental designs into typed, parameterized protocols for liquid handlers, incubators, plate readers, LC-MS, and robotic arms.\n• Build deterministic static-analysis and simulation verifiers enforcing chemical compatibility, biosafety containment rules, volume/dead-volume conservation, thermal limits, and deck collision constraints before any hardware actuation.\n• Design Model Context Protocol (MCP) and typed tool interfaces connecting frontier scientific reasoning agents to LIMS databases, electronic lab notebooks, compound inventories, and live instrument telemetry.\n• Engineer real-time anomaly recovery loops capable of pausing workcells, invoking vision/telemetry diagnostic agents on failed pipetting or phase-separation events, and safely replanning schedules.\n• Implement immutable provenance graphs linking every physical microliter transfer, environmental sensor reading, and raw assay file directly back to the originating agent hypothesis and model checkpoint.\n• Build high-fidelity digital-twin lab simulators allowing hundreds of end-to-end multi-day autonomous discovery campaigns to be stress-tested in CI every night.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, TypeScript, Rust/Go, Temporal, FastAPI/gRPC, PostgreSQL, TimescaleDB/ClickHouse, MCP/agent harnesses, PyLabRobot/SiLA2 hardware interfaces, Kubernetes, AWS. Workplace: Hybrid in Boston / Cambridge, MA. Compensation: $195,000 – $265,000 base salary plus generous equity, comprehensive benefits, and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of software systems, robotics orchestration, or applied AI agent engineering experience building mission-critical stateful execution platforms.\n• Strong production mastery of Python and TypeScript, Rust, or Go, including durable workflow engines (Temporal, Dagster, or custom state machines) and event-driven telemetry pipelines.\n• Proven experience building reliable LLM agent harnesses—featuring strict schema validation, deterministic guardrails, formal/rule-based pre-execution verifiers, and replayable audit traces.\n• Solid grasp of scheduling optimization (job-shop / constraint programming), hardware state machines, and fault-tolerant hardware-in-the-loop testing.\n• Track record owning safety-critical or high-consequence software where bugs affect physical equipment or regulated workflows.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience with laboratory automation ecosystems (Hamilton, Tecan, Opentrons, PyLabRobot, SiLA 2, LIMS, robotic workcell schedulers) or industrial robotics middleware (ROS2).\n• Familiarity with constraint solvers (OR-Tools, Z3) for temporal lab scheduling and resource deadlock prevention.\n• Background in cheminformatics, bioinformatics data pipelines, or scientific instrument drivers.",
      "updated_at": "2026-11-15T17:42:37.080Z",
      "created_at": "2026-11-15T16:52:19.295Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ai-agent-autonomous-lab-orchestration",
        "markdown": "https://jobs.talent247.co/jobs/senior-ai-agent-autonomous-lab-orchestration.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ai-agent-autonomous-lab-orchestration/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ai-agent-autonomous-lab-orchestration/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "15d29e32-06e6-4231-873f-8efc552cbe6e",
      "slug": "staff-machine-learning-engineer-agent-evaluation",
      "title": "Staff Machine Learning Engineer, Agent Evaluation & Judges",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 235000,
        "max": 330000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$235k – $330k",
        "subtitle": "annual package"
      },
      "summary": "Staff Machine Learning Engineer, Agent Evaluation & Judges (Machine Learning) — San Francisco, CA · Hybrid · $235k – $330k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally scaled data and artificial intelligence platform leader headquartered in San Francisco, empowering thousands of enterprises to build, evaluate, govern, and serve custom agentic AI systems over their proprietary data estates. Their AI Research & Agent Evaluation organization builds the automated judge models, synthetic benchmark generators, and reinforcement-learning feedback loops that allow enterprise engineering teams to hill-climb compound agent quality systematically without manual labeling bottlenecks.\n\nThey have retained Talent247 to appoint a Staff Machine Learning Engineer for Agent Evaluation & Judges in San Francisco, CA. Combining post-training research depth with large-scale product impact, the successful candidate will architect custom generative judge models, automatic prompt/weight optimizers, and trajectory-diagnosis algorithms deployed across tens of thousands of enterprise AI pipelines.\n\nKEY RESPONSIBILITIES\n• Train and post-train specialized, bias-mitigated generative LLM-as-a-judge and process-reward models that evaluate multi-turn tool-calling trajectories, retrieval groundedness, SQL/code correctness, and domain policy compliance.\n• Eliminate classic judge failure modes—position bias, verbosity bias, self-enhancement bias, and rubric drift—through principled preference alignment, chain-of-critique post-training, and small-sample human calibration adapts.\n• Pioneer automated synthetic test-set synthesis and adversarial edge-case mining pipelines generating realistic multi-hop evaluation benchmarks directly from enterprise document/table schemas and production trace logs.\n• Co-design closed-loop agent optimization algorithms (automated prompt/rubric evolution, teacher-student distillation, and online RLAIF/GRPO) that turn evaluation failures directly into measurable model upgrades.\n• Partner with core platform engineering teams to ship sub-100ms streaming guardrail judges and asynchronous batch trace evaluators inside unified developer SDKs and observability UIs.\n• Publish influential research and open benchmarks advancing reliable evaluation of enterprise compound AI systems.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, FSDP/Megatron, Verl/TRL, vLLM/SGLang, Ray, Spark, OpenTelemetry agent tracing, AWS/GCP/Azure GPU clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $235,000 – $330,000 base salary plus substantial equity grants, annual refreshers/bonus, and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of machine learning engineering or applied AI research experience, including Staff-level technical leadership in LLM post-training, evaluation systems, reward modeling, or compound AI optimization.\n• Deep hands-on track record training custom language models (SFT, DPO, RLHF/RLAIF, GRPO) and building statistically rigorous LLM evaluation frameworks in PyTorch.\n• Strong command of psychometric/statistical calibration, inter-rater reliability, active learning from sparse human feedback, and causal error attribution in multi-step agent graphs.\n• Proven ability to translate frontier ML research into widely adopted developer frameworks or enterprise cloud platform products.\n• Track record mentoring senior ML engineers and setting multi-quarter technical strategy across research and product engineering pods.\n\nPREFERRED / BONUS EXPERIENCE\n• Publications at NeurIPS, ICML, ICLR, ACL, or EMNLP on LLM judges, reward modeling, automated prompt/agent optimization (DSPy/TextGrad-class paradigms), or synthetic data generation.\n• Experience optimizing compact 3B–14B judge/verifier models for high-throughput, low-cost production trace scoring.\n• Core contributions to major open-source AI evaluation, tracing, or agent lifecycle frameworks.",
      "updated_at": "2026-11-15T17:42:30.614Z",
      "created_at": "2026-11-15T16:51:15.915Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-machine-learning-engineer-agent-evaluation",
        "markdown": "https://jobs.talent247.co/jobs/staff-machine-learning-engineer-agent-evaluation.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-machine-learning-engineer-agent-evaluation/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-machine-learning-engineer-agent-evaluation/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "a33921a5-86ef-459f-8428-272eb467654e",
      "slug": "staff-neural-retrieval-search-ranking-ml-engineer",
      "title": "Staff Neural Retrieval & Search Ranking ML Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 255000,
        "max": 355000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$255k – $355k",
        "subtitle": "annual package"
      },
      "summary": "Staff Neural Retrieval & Search Ranking ML Engineer (Machine Learning) — San Francisco, CA · On-site · $255k – $355k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar real-time AI answer and deep-research engine headquartered in San Francisco, serving billions of grounded, citation-backed queries every month across consumer, enterprise, and developer APIs. Unlike static pretrained LLMs whose world knowledge freezes at training cutoff, their core technical moat is a planetary-scale, sub-100ms neural web search and multi-stage ranking stack built specifically for LLM reasoning agents rather than ten-blue-link human clicks.\n\nThey have retained Talent247 to appoint a Staff Neural Retrieval & Search Ranking ML Engineer onsite in San Francisco, CA. Serving as a technical lead within their Core Search & Retrieval ML organization, the successful candidate will architect the next generation of dense, learned-sparse, late-interaction, and cross-encoder ranking models over hundreds of billions of live web and enterprise documents.\n\nKEY RESPONSIBILITIES\n• Define the architectural roadmap and training recipes for multi-stage neural search retrieval: multilingual dense embeddings, learned sparse/lexical expansion, token-level late interaction, and ultra-low-latency cross-encoder rerankers.\n• Optimize retrieval objective functions specifically for downstream LLM reasoning utility—maximizing factual completeness, primary-source authority, temporal freshness, and multi-hop evidence diversity rather than clickbait CTR.\n• Co-design quantized embedding formats (binary/INT4/Matryoshka representations) and GPU/CPU approximate nearest-neighbor (HNSW/IVF-PQ/DiskANN) pruning cascades alongside C++/Rust search engine systems engineers to stay under <40ms p99 retrieval budgets.\n• Build continuous distillation loops where heavy offline multi-document LLM synthesis judges generate billions of calibrated hard-negative relevance labels for real-time student rankers.\n• Pioneer query-rewriting, entity/temporal disambiguation, and dynamic sub-index routing models handling breaking news, code docs, financial filings, and scholarly literature.\n• Lead online interleaved experiments and end-to-end answer-quality evaluations impacting tens of millions of daily active users.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, C++20/Rust, CUDA/TensorRT/ONNX, custom petabyte vector & inverted-index search engine, Ray, ClickHouse, multi-node H100/B200 GPU clusters. Workplace: Onsite in San Francisco, CA. Compensation: $255,000 – $355,000 base salary plus high-velocity unicorn equity, daily meals, and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of machine learning engineering and applied research experience in web-scale search ranking, neural information retrieval, or large-scale recommendation/ads retrieval systems.\n• Deep track record training contrastive dense retrievers, SPLADE/learned sparse models, ColBERT-style late-interaction architectures, and distilled cross-encoder rerankers at multi-billion-document scale.\n• Strong systems awareness of ANN index trade-offs (RAM vs. NVMe, quantization recall loss, filtered search, real-time index updates) and sub-10ms GPU/SIMD model inference.\n• Mastery of counterfactual offline IR metrics, LLM-assisted relevance adjudication, and high-velocity online A/B experimentation.\n• Proven Staff-level ownership driving cross-team roadmaps across ML scientists and C++/Rust search systems engineers.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior Staff/Senior tenure on a major commercial web search engine, developer/enterprise search platform, or frontier RAG research team.\n• Publications at SIGIR, WSDM, WWW, ACL, EMNLP, or NeurIPS on neural retrieval, passage ranking, or retrieval-augmented reasoning.\n• Hands-on C++ or Rust familiarity for profiling inference-inside-index scoring kernels.",
      "updated_at": "2026-11-15T17:42:10.561Z",
      "created_at": "2026-11-15T16:51:05.732Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-neural-retrieval-search-ranking-ml-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-neural-retrieval-search-ranking-ml-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-neural-retrieval-search-ranking-ml-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-neural-retrieval-search-ranking-ml-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "80f268f8-f4ba-4179-bfd1-596b829f3d77",
      "slug": "applied-ml-infrastructure-python-sdk-engineer",
      "title": "Applied ML Infrastructure & Python SDK Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 195000,
        "max": 275000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$195k – $275k",
        "subtitle": "annual package"
      },
      "summary": "Applied ML Infrastructure & Python SDK Engineer (Applied AI) — San Francisco, CA · Hybrid · $195k – $275k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a developer-beloved serverless GPU and AI infrastructure platform with engineering hubs in New York and San Francisco. Their core thesis is that AI engineers should define distributed GPU clusters, microVM code sandboxes, scheduled batch pipelines, and low-latency inference endpoints entirely in idiomatic Python—with zero YAML, zero Terraform, and sub-second feedback loops.\n\nThey have engaged Talent247 to recruit an Applied ML Infrastructure & Python SDK Engineer in San Francisco, CA. Sitting directly at the boundary between their Rust serverless compute engine and the global ML developer community, the successful candidate will shape the Python SDK primitives, distributed RL/training abstractions, and flagship reference architectures adopted by thousands of AI teams.\n\nKEY RESPONSIBILITIES\n• Design and evolve ergonomic, rock-solid Python SDK primitives for declarative container environments, GPU memory snapshotting, distributed multi-node gang scheduling, Ephemeral Sandboxes, and high-concurrency async RPCs.\n• Architect production-grade reference implementations and reusable building blocks for frontier ML workloads—including disaggregated vLLM/SGLang serving, distributed RLVR rollout gyms, petabyte video/data curation, and biomolecular folding pipelines.\n• Partner directly with ML researchers and Staff engineers at flagship AI startups in San Francisco to debug gnarly distributed PyTorch, NCCL, CUDA Graph, and async concurrency edge cases.\n• Optimize Python client-runtime serialization, gRPC/HTTP2 streaming throughput, asyncio event-loop overhead, and introspectable local-to-cloud stack traces.\n• Build automated end-to-end performance benchmarks tracking cold-start, throughput, and multi-node scaling across every new open-weight model family and GPU generation (H100/H200/B200).\n• Author crystal-clear technical guides, architectural deep-dives, and open-source blueprints that set the standard for developer-centric AI infrastructure.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python (deep internals: AST/bytecode, `asyncio`, typing, C/Rust PyO3 extensions), Rust, gRPC/Protobuf, PyTorch, vLLM/SGLang, TensorRT-LLM, CUDA, distributed sandboxes. Workplace: Hybrid in San Francisco, CA. Compensation: $195,000 – $275,000 base salary plus high-upside equity grants and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of ML infrastructure, developer SDK/framework, or applied distributed systems engineering experience.\n• Deep mastery of Python internals—including `asyncio` concurrency, import/serialization mechanics, type-system ergonomics, and CPython/Rust (`PyO3`) boundary performance.\n• Hands-on production experience running modern GPU workloads: LLM inference engines (vLLM/SGLang), distributed PyTorch training/fine-tuning, or containerized code-execution sandboxes.\n• Strong API design taste: ability to distill complex distributed systems semantics into intuitive, composable abstractions that developers love.\n• High autonomy and crisp written technical communication skills.\n\nPREFERRED / BONUS EXPERIENCE\n• Core authorship or maintenance of widely used Python ML, async networking, workflow-orchestration, or developer-tooling libraries.\n• Working fluency in Rust or C++ to trace performance across the SDK-to-runtime boundary.\n• Experience building RL rollout infrastructure, multimodal video/audio pipelines, or AI coding-agent sandboxes.",
      "updated_at": "2026-11-15T17:41:58.812Z",
      "created_at": "2026-11-15T16:51:02.211Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/applied-ml-infrastructure-python-sdk-engineer",
        "markdown": "https://jobs.talent247.co/jobs/applied-ml-infrastructure-python-sdk-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/applied-ml-infrastructure-python-sdk-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/applied-ml-infrastructure-python-sdk-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8c1a6753-41f0-4a2d-9717-dcb06a0a76e0",
      "slug": "senior-rust-systems-engineer-serverless-gpu",
      "title": "Senior Rust Systems Engineer, Serverless GPU Runtime",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "New York, NY",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 230000,
        "max": 330000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$230k – $330k",
        "subtitle": "annual package"
      },
      "summary": "Senior Rust Systems Engineer, Serverless GPU Runtime (AI Infrastructure) — New York, NY · On-site · $230k – $330k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an acclaimed serverless AI and data cloud headquartered in New York City, revered by ML engineers and frontier agent builders for eliminating YAML, Dockerfiles, and multi-minute GPU boot times. Behind their effortless Python interface sits a ground-up custom Rust hypervisor, container runtime, and memory-snapshotting stack built from first principles to boot isolated GPU containers and untrusted code sandboxes in hundreds of milliseconds.\n\nThey have retained Talent247 to appoint a Senior Rust Systems Engineer for their Serverless GPU Runtime onsite in New York, NY. Working in a compact, deeply technical systems engineering culture, the successful candidate will push Linux kernel, hypervisor, and CUDA driver virtualization boundaries to make 10,000-container GPU scaleouts feel instantaneous.\n\nKEY RESPONSIBILITIES\n• Architect and harden custom container and microVM sandboxing runtimes in Rust utilizing Linux namespaces, cgroups v2, seccomp, gVisor/KVM primitives, and hardware-enforced multi-tenant isolation.\n• Engineer CPU + NVIDIA GPU memory checkpoint-and-restore snapshotting pipelines restoring fully initialized PyTorch/CUDA contexts and multi-gigabyte VRAM states in <500ms.\n• Build lazy-loading FUSE / `io_uring` filesystem clients and zero-copy page-cache layers streaming only the exact Python bytecode and shared-library pages touched during cold start.\n• Design high-density bin-packing host agents and low-jitter virtual networking stacks (`eBPF`, `nftables`, WireGuard, custom HTTP/gRPC multiplexers) across global GPU bare-metal and cloud fleets.\n• Collaborate upstream/downstream around NVIDIA GPU driver internals (`nvidia-uvm`, CUDA Driver API, MIG/MPS, NVLink partitioning) to eliminate driver initialization bottlenecks.\n• Maintain extreme reliability and security rigor across sandboxes executing millions of untrusted LLM-generated code and RL rollout workloads daily.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust (async Tokio, `nix`/`libc`, `io_uring`), C, Linux kernel internals (KVM, cgroups v2, FUSE, `userfaultfd`, eBPF), CUDA Driver API, bare-metal & multi-cloud H100/B200 GPU hosts. Workplace: Onsite in New York, NY (SoHo/Downtown). Compensation: $230,000 – $330,000 base salary plus premier high-upside equity, comprehensive healthcare, daily catered lunches, and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of low-level systems engineering experience building operating-system runtimes, hypervisors/containers, storage engines, or high-performance networking infrastructure in Rust, C, or C++.\n• Deep fluency in Rust systems programming alongside intimate knowledge of Linux kernel mechanics (virtual memory, page faults, syscalls, namespaces, cgroups, filesystems, async IO).\n• Track record diagnosing subtle kernel/userspace concurrency bugs, tail-latency stalls, memory leaks, and security isolation edge cases in production.\n• First-principles engineering taste—comfortable reading kernel or driver source code when off-the-shelf container tooling hits architectural walls.\n• Excitement to collaborate onsite in NYC with fellow systems, storage, and compiler engineers.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience with CRIU, Firecracker/Cloud Hypervisor, gVisor, CUDA checkpoint/restore (`cuda-checkpoint`), or GPU virtualization.\n• Familiarity with FUSE/`io_uring` high-throughput filesystems or eBPF packet/syscall filtering.\n• Core contributions to Rust systems crates, Linux kernel subsystems, or container runtimes (`youki`, `containerd`, `runc`).",
      "updated_at": "2026-11-15T17:41:36.023Z",
      "created_at": "2026-11-15T16:50:35.701Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-rust-systems-engineer-serverless-gpu",
        "markdown": "https://jobs.talent247.co/jobs/senior-rust-systems-engineer-serverless-gpu.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-rust-systems-engineer-serverless-gpu/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-rust-systems-engineer-serverless-gpu/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8941465d-bd0c-492b-807d-63e25f451f1f",
      "slug": "ml-solutions-forward-deployed-inference-engineer",
      "title": "ML Solutions & Forward-Deployed Inference Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 190000,
        "max": 265000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$190k – $265k",
        "subtitle": "annual package"
      },
      "summary": "ML Solutions & Forward-Deployed Inference Engineer (Applied AI) — San Francisco, CA · Hybrid · $190k – $265k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco AI inference platform trusted by category-defining voice AI, coding assistant, healthcare, and multimodal media companies to run custom models in production at massive scale. Their ML Solutions & Forward-Deployed Inference team acts as the tip of the spear—partnering directly with founders and Staff ML engineers at breakout AI companies to take custom PyTorch checkpoints from research notebooks to sub-100ms global production endpoints.\n\nThey have engaged Talent247 to recruit an ML Solutions & Forward-Deployed Inference Engineer in San Francisco, CA. Blending hands-on GPU model optimization with high-velocity customer architecture, the appointed engineer will solve gnarly real-world serving bottlenecks across LLMs, streaming audio, vision-language, and diffusion pipelines.\n\nKEY RESPONSIBILITIES\n• Partner directly with customer ML engineering teams during critical launches to package, benchmark, quantize, and deploy custom PyTorch, TensorRT-LLM, SGLang, and vLLM pipelines onto dedicated GPU clusters.\n• Profile end-to-end request lifecycles—from tokenization, audio/image preprocessing, and dynamic batching through GPU kernel execution and WebSocket/gRPC streaming—cutting latency by 2x–5x and GPU cost by 40%+.\n• Implement production-ready compound inference chains combining custom LoRA adapters, Whisper/streaming ASR, expressive TTS, embeddings/rerankers, and speculative decoding draft models.\n• Design realistic load-testing suites matching customer burst concurrency, context-length distributions, and multi-turn prefix-cache patterns prior to production cutover.\n• Build clean open-source reference templates, Python packaging SDK improvements, and automated quantization recipes benefiting thousands of platform developers.\n• Serve as the voice of frontier ML builders internally, shaping product roadmaps across core model performance and multi-cloud infrastructure teams.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, TensorRT-LLM, vLLM, SGLang, Triton Inference Server, Docker/Kubernetes, FastAPI/gRPC/WebSockets, FP8/INT4 quantization toolkits, H100/H200/B200 GPUs. Workplace: Hybrid in San Francisco, CA. Compensation: $190,000 – $265,000 base salary plus meaningful equity grants, performance incentives, and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of ML engineering, inference optimization, or forward-deployed solutions engineering experience deploying deep learning models into latency-sensitive production environments.\n• Strong hands-on proficiency in PyTorch alongside modern LLM/generative serving engines (vLLM, TensorRT-LLM, SGLang, or Diffusers/TorchCompile).\n• Practical command of GPU memory sizing, KV-cache math, continuous batching, tensor parallelism, and quantization trade-offs across L40S, H100, H200, and B200 GPUs.\n• Crisp Python systems engineering skills (asyncio, multiprocessing, shared memory, containerization, OpenTelemetry/Prometheus metrics).\n• Articulate, low-ego collaborator energized by debugging live technical challenges alongside elite AI engineering teams.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience optimizing real-time streaming audio/voice pipelines (TTS/ASR), video/image diffusion models, or custom C++/CUDA PyTorch extensions.\n• Prior forward-deployed, applied AI, or developer-facing ML engineering experience at a high-growth AI infrastructure company.\n• Published technical benchmarks, blog posts, or open-source inference recipes.",
      "updated_at": "2026-11-15T17:41:25.574Z",
      "created_at": "2026-11-15T16:50:35.294Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/ml-solutions-forward-deployed-inference-engineer",
        "markdown": "https://jobs.talent247.co/jobs/ml-solutions-forward-deployed-inference-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/ml-solutions-forward-deployed-inference-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/ml-solutions-forward-deployed-inference-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "aa494718-2584-4675-a745-48f043e44ef0",
      "slug": "senior-multi-cloud-gpu-orchestration-engineer",
      "title": "Senior Multi-Cloud GPU Orchestration Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 200000,
        "max": 280000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$200k – $280k",
        "subtitle": "annual package"
      },
      "summary": "Senior Multi-Cloud GPU Orchestration Engineer (MLOps & Platform) — New York, NY · Hybrid · $200k – $280k. THE CLIENT & MANDATE Our client is a leading AI…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a leading AI inference cloud headquartered in San Francisco with a major distributed infrastructure hub in New York City. To shield fast-scaling enterprise and AI-native customers from single-cloud GPU capacity crunches and regional outages, their Multi-Cloud Compute Fabric orchestrates tens of thousands of GPUs seamlessly across AWS, GCP, Azure, OCI, and specialized bare-metal neoclouds as a single logical inference mesh.\n\nThey have partnered with Talent247 to hire a Senior Multi-Cloud GPU Orchestration Engineer in New York, NY. The successful candidate will design global cross-cloud capacity brokers, sub-second scale-from-zero container/weight loaders, and latency-aware Envoy/Rust traffic routers sustaining 99.995% availability.\n\nKEY RESPONSIBILITIES\n• Architect federated Kubernetes control planes and declarative placement schedulers in Go and Rust brokering real-time H100/H200/B200 GPU capacity across 10+ hyperscale and bare-metal cloud providers.\n• Build global Layer-7 AI traffic routers (Envoy / Rust data planes) balancing geo-proximity, queue depth, prefix-cache locality, and live cross-region GPU health signals with zero-downtime failover.\n• Engineer ultra-fast cross-cloud model weight & container caching layers leveraging NVMe tiering, OCI lazy-pulling, and high-throughput regional object stores to cut replica cold starts to single-digit seconds.\n• Design secure dedicated-VPC customer data planes (PrivateLink, Private Service Connect, WireGuard/Cilium mesh, customer-managed KMS) meeting SOC2 Type II and HIPAA enterprise isolation requirements.\n• Implement predictive autoscaling and spot/on-demand/reserved GPU arbitrage engines optimizing multi-million-dollar monthly compute fleets.\n• Own chaos-engineering game days simulating full cloud-region blackouts and GPU hardware degradations without dropping streaming tokens.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Go, Rust, Python, Kubernetes (custom operators, Karpenter/custom provisioners), Cilium/eBPF, Envoy, gRPC, Terraform/Pulumi, Prometheus/ClickHouse, AWS/GCP/Azure/OCI & bare-metal clouds. Workplace: Hybrid in New York, NY. Compensation: $200,000 – $280,000 base salary plus generous equity grants and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of cloud infrastructure, distributed systems, or multi-cluster Kubernetes platform engineering experience in Go or Rust.\n• Deep production mastery of Kubernetes federation, custom controllers/operators, CNI networking (Cilium/Calico/WireGuard), and high-throughput L7 proxies (Envoy).\n• Hands-on experience managing production GPU fleets across multiple cloud providers, including NVIDIA driver/device-plugin lifecycles, MIG, and GPU health monitoring (DCGM).\n• Strong grounding in cloud security primitives (IAM Federation, PrivateLink/PSC, mTLS service meshes, zero-trust tenant isolation).\n• Track record owning mission-critical data-plane or control-plane infrastructure under strict 99.99%+ uptime SLAs.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience building serverless GPU autoscalers, scale-to-zero cold-start accelerators, or multi-node InfiniBand/RoCE gang scheduling across neoclouds.\n• Contributions to Envoy AI Gateway, Kubernetes Cluster API, Cilium, or OCI image-streaming runtimes.\n• Familiarity with LLM serving patterns (disaggregated prefill/decode routing, KV-cache session affinity).",
      "updated_at": "2026-11-15T17:41:12.854Z",
      "created_at": "2026-11-15T16:50:31.227Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-multi-cloud-gpu-orchestration-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-multi-cloud-gpu-orchestration-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-multi-cloud-gpu-orchestration-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-multi-cloud-gpu-orchestration-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "39558187-746d-4d0a-8611-76a91c24a588",
      "slug": "staff-model-performance-tensorrt-llm-engineer",
      "title": "Staff Model Performance & TensorRT-LLM Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 240000,
        "max": 335000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$240k – $335k",
        "subtitle": "annual package"
      },
      "summary": "Staff Model Performance & TensorRT-LLM Engineer (AI Infrastructure) — San Francisco, CA · Hybrid · $240k – $335k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar AI inference infrastructure platform headquartered in San Francisco, powering mission-critical low-latency model serving for breakout AI-native applications and healthcare, voice, code, and creative-media enterprises. Rather than offering generic shared endpoints, their Model Performance Engineering group squeezes hardware-limit throughput and sub-50ms time-to-first-token out of custom multimodal, audio, diffusion, and 400B+ LLM architectures across dedicated Hopper and Blackwell clusters.\n\nThey have retained Talent247 to appoint a Staff Model Performance & TensorRT-LLM Engineer in San Francisco, CA. Operating at the intersection of GPU kernel optimization, compiler runtimes, and speculative decoding research, the successful candidate will own flagship model optimization benchmarks and core runtime primitives across their inference fleet.\n\nKEY RESPONSIBILITIES\n• Architect end-to-end TensorRT-LLM, SGLang, and vLLM runtime optimizations across NVIDIA H100, H200, B200, and GB200 NVL72 clusters—delivering industry-leading latency-per-dollar Pareto frontiers.\n• Design and tune low-bit quantization pipelines (NVFP4, FP8 KV-cache & GEMMs, INT4 AWQ/GPTQ) with custom calibration recipes preserving lossless reasoning, code synthesis, and audio prosody quality.\n• Build online-adaptive speculative decoding engines (EAGLE-style draft heads, n-gram/PLD lookup, multi-token prediction) tailored dynamically to live customer domain traffic.\n• Author custom CUDA, CUTLASS, and Triton kernels for non-standard attention masks, multimodal vision/audio encoders, Diffusion Transformers (DiTs), and disaggregated prefill-decode KV transfers.\n• Diagnose deep GPU runtime bottlenecks across CUDA Graph capture, Python/C++ host launch overhead, NCCL/NVSHMEM collectives, and HBM memory fragmentation using Nsight Compute and Nsight Systems.\n• Publish authoritative technical deep-dives and mentor senior inference engineers across core runtime and forward-deployed performance pods.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: C++20, CUDA, CUTLASS/CuTe, Triton, Python, TensorRT-LLM, SGLang, vLLM, FlashInfer, PyTorch, Nsight Compute/Systems, Hopper & Blackwell GPU pods. Workplace: Hybrid in San Francisco, CA. Compensation: $240,000 – $335,000 base salary plus high-upside Series-stage equity, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 6+ years of GPU inference performance engineering, deep learning compiler/runtime development, or HPC kernel optimization experience.\n• Deep production mastery of TensorRT-LLM, SGLang, or vLLM internals—including custom C++/CUDA plugins, paged attention, chunked prefill, and multi-GPU tensor/expert parallelism.\n• Proven track record profiling and eliminating kernel-level and host-bound bottlenecks with NVIDIA Nsight Systems and Nsight Compute.\n• Hands-on expertise in post-training quantization (FP8, NVFP4, INT4/INT8) and speculative decoding architectures.\n• Strong communication and technical leadership skills driving cross-team runtime architecture decisions.\n\nPREFERRED / BONUS EXPERIENCE\n• Core upstream contributions to TensorRT-LLM, FlashInfer, SGLang, vLLM, CUTLASS, or xDiT.\n• Experience optimizing non-LLM generative workloads (streaming TTS/ASR audio models, real-time video DiTs, or large cross-encoder rerankers) alongside frontier LLMs.\n• Direct hands-on tuning experience on NVIDIA Blackwell B200 / GB200 NVL72 hardware.",
      "updated_at": "2026-11-15T17:40:58.348Z",
      "created_at": "2026-11-15T16:50:27.388Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-model-performance-tensorrt-llm-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-model-performance-tensorrt-llm-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-model-performance-tensorrt-llm-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-model-performance-tensorrt-llm-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "41dd7aeb-577f-4f0d-9430-e194765f8a6a",
      "slug": "senior-full-stack-engineer-ai-developer-cloud",
      "title": "Senior Full-Stack Engineer, AI Developer Cloud",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 190000,
        "max": 265000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$190k – $265k",
        "subtitle": "annual package"
      },
      "summary": "Senior Full-Stack Engineer, AI Developer Cloud (Applied AI) — San Francisco, CA · Hybrid · $190k – $265k. THE CLIENT & MANDATE Our client is a premier…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco AI Acceleration Cloud serving hundreds of thousands of AI developers, research teams, and enterprise engineering orgs with instant serverless inference, fine-tuning/RL pipelines, and dedicated GPU clusters. Their AI Developer Cloud product team crafts the interactive model playgrounds, side-by-side evaluation workbenches, fine-tuning telemetry consoles, and sub-millisecond API billing planes that turn raw GPU compute into an effortless developer experience.\n\nThey have retained Talent247 to appoint a Senior Full-Stack Engineer for their AI Developer Cloud in San Francisco, CA. Owning end-to-end product surfaces from reactive TypeScript/React interfaces down to high-throughput Go and ClickHouse API services, the successful candidate will build developer tools used daily by the world's top AI builders.\n\nKEY RESPONSIBILITIES\n• Architect and ship snappy, keyboard-first developer workflows in TypeScript, React/Next.js, and Tailwind—including multi-model streaming playgrounds, structured-output schema builders, RL/LoRA loss-curve explorers, and GPU cluster topology dashboards.\n• Build high-concurrency Go and TypeScript backend control-plane APIs handling API key provisioning, scoped RBAC organizations, dedicated endpoint deployments, and batch/fine-tuning job lifecycle orchestration.\n• Engineer real-time token-usage metering, per-second GPU billing ledgers, credit/quota enforcement, and sub-second ClickHouse observability analytics (TTFT, tok/s, cache hit rates, error distributions) over billions of daily API events.\n• Design seamless developer onboarding, interactive SDK code generators, and OpenTelemetry trace viewers for debugging multi-turn agentic tool calls.\n• Partner directly with inference runtime and research teams to launch zero-day UI and SDK experiences alongside major open-weight text, vision, audio, and image/video model releases.\n• Uphold rigorous frontend performance, API schema ergonomics, and security standards across public cloud and VPC-hosted enterprise portals.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: TypeScript, React, Next.js, Go, Python, PostgreSQL, ClickHouse, Redis, Kafka, Stripe/usage-billing ledgers, Kubernetes, AWS/GCP. Workplace: Hybrid in San Francisco, CA. Compensation: $190,000 – $265,000 base salary plus meaningful unicorn equity, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of full-stack software engineering experience building developer platforms, cloud consoles, observability tools, or high-scale B2B SaaS/API products.\n• Mastery of modern TypeScript, React, and Next.js—with deep care for streaming UI responsiveness, complex real-time data visualization, and clean component systems.\n• Strong backend systems engineering skills in Go, TypeScript/Node, Rust, or Python backed by PostgreSQL, Redis, and analytical stores such as ClickHouse.\n• Experience designing developer-loved REST/SSE/WebSocket APIs, authentication/RBAC models, or usage-based metering pipelines.\n• High product velocity and taste—capable of taking an ambiguous developer workflow from sketch to polished production launch in days.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience at an AI platform, cloud infrastructure provider, observability company, or developer-tools startup.\n• Hands-on familiarity with LLM API primitives (streaming SSE deltas, tool calling, structured JSON schemas, fine-tuning datasets, tokenizers).\n• Open-source SDK, CLI, or UI library authorship.",
      "updated_at": "2026-11-15T17:39:59.923Z",
      "created_at": "2026-11-15T16:50:23.996Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-full-stack-engineer-ai-developer-cloud",
        "markdown": "https://jobs.talent247.co/jobs/senior-full-stack-engineer-ai-developer-cloud.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-full-stack-engineer-ai-developer-cloud/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-full-stack-engineer-ai-developer-cloud/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "a3e80361-603e-4f8f-a81c-4b5382987c23",
      "slug": "senior-gpu-cluster-rdma-networking-engineer",
      "title": "Senior GPU Cluster & RDMA Networking Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 225000,
        "max": 315000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $315k",
        "subtitle": "annual package"
      },
      "summary": "Senior GPU Cluster & RDMA Networking Engineer (AI Infrastructure) — San Francisco, CA · Hybrid · $225k – $315k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar AI Acceleration Cloud headquartered in San Francisco operating massive bare-metal clusters of NVIDIA H100, H200, B200, and GB200 NVL72 superpods across North America and Europe. Delivering hyperscale-beating distributed training and disaggregated MoE inference performance depends on lossless, microsecond-tuned 400G/800G InfiniBand and RoCEv2 RDMA fabrics engineered down to the NIC register and NCCL transport layer.\n\nThey have engaged Talent247 to hire a Senior GPU Cluster & RDMA Networking Engineer in San Francisco, CA. Joining their Bare-Metal Supercomputing Infrastructure group, the successful candidate will architect rail-optimized AI fabrics, tune NCCL/NVSHMEM collective transports, and build real-time RDMA telemetry across tens of thousands of GPUs.\n\nKEY RESPONSIBILITIES\n• Architect, deploy, and scale non-blocking rail-optimized 400G and 800G NVIDIA Quantum-2/X800 InfiniBand and Spectrum-X / Broadcom Tomahawk5 RoCEv2 fabrics for multi-thousand-GPU clusters.\n• Co-optimize NCCL, NVSHMEM, and libfabric/UCX collective communication parameters—tuning congestion control (DCQCN, ECN, PFC, adaptive routing), QP scaling, and PCIe/NVSwitch topology mappings.\n• Build automated pre-flight fabric qualification suites (`ib_write_bw`, `nccl-tests`, bisectional all-reduce/all-to-all sweeps) and continuous link-flap/BER optical degradation detectors that quarantine degraded rails before training jobs stall.\n• Engineer secure multi-tenant RDMA isolation (InfiniBand PKeys, EVPN-VXLAN / SRv6 + RoCEv2 ACLs, ConnectX-7/8 BlueField-3 DPU offloads) for dedicated enterprise training pods.\n• Partner directly with inference and training kernel engineers to optimize low-latency internode expert-parallel all-to-all dispatch and KV-cache RDMA transfers.\n• Drive vendor silicon/firmware root-cause debugging across switches, optical transceivers, DPUs, and Linux RDMA (`rdma-core` / MLNX_OFED / DOCA) kernel stacks.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: 400G/800G InfiniBand (UFM) & RoCEv2 (SONiC / EOS / Cumulus), ConnectX-7/8 & BlueField-3 DPUs, GPUDirect RDMA, NCCL/NVSHMEM, Linux networking, Python/Go/Rust automation, Prometheus/eBPF. Workplace: Hybrid in San Francisco, CA. Compensation: $225,000 – $315,000 base salary plus substantial equity ownership and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of high-performance datacenter networking, HPC interconnect, or AI GPU fabric engineering experience.\n• Deep hands-on mastery of InfiniBand and/or lossless Ethernet RoCEv2 at 200G/400G/800G scale—including BGP CLOS fabrics, PFC/ECN/DCQCN congestion control, adaptive routing, and optical link physics.\n• Strong command of GPUDirect RDMA, PCIe root-complex / ACS / IOMMU interactions, `rdma-core` / DOCA / OFED drivers, and `nccl-tests` performance profiling.\n• Proficient infrastructure software skills in Python, Go, or Rust for switch/fabric automation, zero-touch provisioning, and streaming telemetry.\n• Proven ability to isolate microsecond tail-latency anomalies across multi-rack synchronous AI workloads.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience bringing up NVIDIA GB200 NVL72 clusters, Spectrum-X Ethernet fabrics, or multi-tenant BlueField-3 DPU virtualization.\n• Familiarity with custom NCCL net-plugins, NVSHMEM internode MoE dispatch, or Ultra Ethernet Consortium (UEC) transports.\n• Prior tenure building AI superclusters at a hyperscaler, neocloud, frontier AI lab, or national supercomputing facility.",
      "updated_at": "2026-11-15T17:39:59.170Z",
      "created_at": "2026-11-15T16:50:20.488Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-gpu-cluster-rdma-networking-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-gpu-cluster-rdma-networking-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-gpu-cluster-rdma-networking-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-gpu-cluster-rdma-networking-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "f1062fb8-e6cc-4c04-8b9b-8a6461ce8d94",
      "slug": "research-scientist-hardware-aware-model",
      "title": "Research Scientist, Hardware-Aware Model Architectures",
      "department": "AI Research",
      "department_code": "RES",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 255000,
        "max": 360000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$255k – $360k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Hardware-Aware Model Architectures (AI Research) — San Francisco, CA · Hybrid · $255k – $360k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an open-science-rooted AI Acceleration Cloud and frontier systems research lab headquartered in San Francisco, renowned globally for originating hardware-aware attention algorithms, state-space / hybrid sequence models, and ultra-fast distributed inference and pre-training stacks. Their Research group bridges theoretical algorithm design directly with GPU SRAM/tensor-core physics so new model architectures train and serve orders of magnitude faster on real silicon.\n\nThey have partnered with Talent247 to appoint a Research Scientist in Hardware-Aware Model Architectures in San Francisco, CA. Working alongside celebrated systems-ML researchers, the successful candidate will invent sub-quadratic sequence models, sparse/hybrid architectures, and low-bit training paradigms validated at multi-billion-parameter scale on dedicated H100/B200 superclusters.\n\nKEY RESPONSIBILITIES\n• Originate novel hardware-aware neural architectures—including hybrid linear/state-space + softmax attention, dynamic sparse MoE routing, latent KV compression, and multi-token prediction heads—co-designed with GPU memory hierarchies.\n• Co-write custom CUDA, CUTLASS/CuTe, or Triton forward-and-backward training kernels ensuring theoretical FLOP reductions translate directly into wall-clock training and inference speedups.\n• Investigate native low-precision pre-training and post-training algorithms (FP8, MXFP4/NVFP4, quantization-aware architecture scaling laws) eliminating outlier instability at scale.\n• Design architecture conversion and distillation recipes capable of retrofitting existing pretrained transformer checkpoints into fast hybrid/sub-quadratic variants using minimal continued pre-training tokens.\n• Lead multi-node scaling-law experiments from 1B ablations through 70B+ reference runs across long-context (1M+ token) reasoning, coding, and multimodal tasks.\n• Publish landmark papers at NeurIPS, ICML, ICLR, and MLSys while releasing high-impact open-source kernels and model implementations.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, CUDA / C++20, CUTLASS 3.x / CuTe, Triton, Megatron/TorchTitan, multi-thousand H100/H200/B200 InfiniBand GPU clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $255,000 – $360,000 base salary plus substantial unicorn equity, research publication freedom, and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Computer Science, Machine Learning, Electrical Engineering, or equivalent premier research track record at the intersection of deep learning architectures and GPU systems.\n• Strong first-author publication record at NeurIPS, ICML, ICLR, MLSys, or OSDI/ASPLOS in efficient sequence modeling, attention algorithms, MoE, quantization, or IO-aware deep learning.\n• Rare dual fluency: capable of both deriving mathematically sound sequence-modeling/optimization dynamics and implementing fast Triton or CUDA/CUTLASS kernels in PyTorch.\n• Proven experience running and debugging multi-node pre-training or continued-pre-training experiments on large language models.\n• Collaborative, open-research ethos excited to push the state of the art across both published science and commercial cloud workloads.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct research contributions to FlashAttention-class IO-aware kernels, Mamba / linear attention / state-space hybrids, sparse attention, or FP8/FP4 training stability.\n• Mastery of NVIDIA Hopper/Blackwell tensor-core instructions, warp specialization, and asynchronous TMA pipelines.\n• Widely adopted open-source ML systems libraries on GitHub.",
      "updated_at": "2026-11-15T17:39:58.424Z",
      "created_at": "2026-11-15T16:50:16.845Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-hardware-aware-model",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-hardware-aware-model.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-hardware-aware-model/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-hardware-aware-model/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "ba337fed-86ad-47eb-b177-b4d7e997f91e",
      "slug": "senior-gpu-control-plane-kubernetes-engineer",
      "title": "Senior GPU Control Plane & Kubernetes Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Redwood City, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 215000,
        "max": 295000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$215k – $295k",
        "subtitle": "annual package"
      },
      "summary": "Senior GPU Control Plane & Kubernetes Engineer (MLOps & Platform) — Redwood City, CA · Hybrid · $215k – $295k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar AI inference and post-training cloud in Redwood City, CA operating tens of thousands of H100, H200, and B200 GPUs across global cloud and bare-metal regions. Orchestrating multi-node Mixture-of-Experts replicas, serverless burst autoscaling, and multi-hundred-gigabyte weight streaming with sub-second cold starts requires a purpose-built Kubernetes GPU control plane far beyond standard cloud autoscalers.\n\nThey have retained Talent247 to hire a Senior GPU Control Plane & Kubernetes Engineer in Redwood City, CA. Joining their Core Compute Infrastructure team, the successful candidate will architect custom Go and Rust Kubernetes operators, topology-aware multi-cluster schedulers, and P2P/RDMA model-weight distribution engines.\n\nKEY RESPONSIBILITIES\n• Design and scale custom Kubernetes controllers, operators, and scheduler plugins in Go and Rust orchestrating multi-node GPU inference pods (NVLink/InfiniBand gang-scheduled replicas) across 20+ global regions.\n• Engineer ultra-fast GPU replica cold-start pipelines—combining pre-warmed container pools, P2P/RDMA peer-to-peer weight broadcasting, and direct NVMe-to-GPU (`cuFile`/GPUDirect Storage) weight loading—bringing 100B+ models online in seconds.\n• Build predictive, queue-depth- and KV-cache-pressure-aware autoscalers seamlessly bridging serverless multi-tenant GPU pools and dedicated enterprise reservations.\n• Implement zero-downtime rolling upgrades, live GPU health telemetry (DCGM/Xid/ECC/link-flap detection), and automated node replacement across heterogeneous cloud and neocloud providers.\n• Architect control-plane isolation, multi-tenant network policies (Cilium/eBPF), and high-throughput global Envoy/Rust ingress routing.\n• Drive cluster bin-packing efficiency and spot/reserved capacity brokering delivering double-digit millions in annual GPU infrastructure savings.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Go, Rust, Kubernetes (client-go, controller-runtime, custom schedulers, DRA), Cilium/eBPF, Envoy, GPUDirect Storage / RDMA, Prometheus/VictoriaMetrics, AWS/GCP/OCI & bare-metal GPU clouds. Workplace: Hybrid in Redwood City, CA. Compensation: $215,000 – $295,000 base salary plus high-value unicorn equity and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of cloud infrastructure, distributed control-plane, or Kubernetes platform engineering experience in Go or Rust.\n• Deep production mastery of Kubernetes internals: writing custom CRDs/controllers, custom scheduler plugins, pod lifecycle hooks, and CNI/CSI/device-plugin integrations at multi-thousand-node scale.\n• Hands-on experience operating NVIDIA GPU clusters in production (MIG/MPS, NVLink/InfiniBand topology awareness, DCGM telemetry, container toolkit internals).\n• Strong systems fundamentals in Linux networking, high-throughput artifact/image distribution, and distributed consensus (etcd/Raft).\n• Proven operational ownership of 99.99% uptime customer-facing cloud control planes.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience optimizing multi-hundred-gigabyte LLM weight loading or container image streaming (Dragonfly, Kraken, Stargz/SOCI, P2P RDMA caches).\n• Experience orchestrating multi-node disaggregated inference (prefill/decode pools) or gang-scheduled LeaderWorkerSet (LWS) workloads.\n• Upstream contributions to Kubernetes, LeaderWorkerSet, Kueue, Cilium, or CNCF projects.",
      "updated_at": "2026-11-15T17:39:47.757Z",
      "created_at": "2026-11-15T16:50:13.456Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-gpu-control-plane-kubernetes-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-gpu-control-plane-kubernetes-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-gpu-control-plane-kubernetes-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-gpu-control-plane-kubernetes-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "38b38403-95bf-44fb-b2a4-fc7ac8ece9fd",
      "slug": "machine-learning-engineer-model-routing-post",
      "title": "Machine Learning Engineer, Model Routing & Post-Training",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Redwood City, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 200000,
        "max": 285000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$200k – $285k",
        "subtitle": "annual package"
      },
      "summary": "Machine Learning Engineer, Model Routing & Post-Training (Machine Learning) — Redwood City, CA · Hybrid · $200k – $285k. THE CLIENT & MANDATE Our client…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier Silicon Valley generative AI inference and model-customization platform in Redwood City, CA enabling developers and enterprises to move from monolithic, high-cost frontier API calls to compound systems of fast, specialized open-weight models. Their Model Post-Training & Adaptive Routing group builds self-serve reinforcement fine-tuning pipelines and sub-millisecond query routers that match or beat frontier quality at a fraction of the latency and cost.\n\nThey have partnered with Talent247 to hire a Machine Learning Engineer for Model Routing & Post-Training in Redwood City, CA. Combining applied RL research with production ML systems engineering, the appointed engineer will build automated distillation, RFT/GRPO post-training recipes, and learned difficulty routers deployed across thousands of production workloads.\n\nKEY RESPONSIBILITIES\n• Design and train lightweight semantic and complexity-aware router models (<5ms overhead) that dynamically dispatch prompts, tool calls, and reasoning steps across specialized LoRA adapters, compact 8B–32B models, and 400B+ MoE backbones.\n• Build turnkey, automated post-training recipes—spanning teacher-to-student trajectory distillation, DPO, and reinforcement fine-tuning (GRPO/RLOO)—that adapt open-weight models from small customer evaluation sets.\n• Develop calibrated speculative-cascade and confidence-abstention mechanisms allowing fast draft/small models to escalate hard reasoning queries seamlessly without accuracy loss.\n• Train custom speculative draft heads (EAGLE / Medusa-style) co-adapted to customer traffic distributions to unlock 2x–3x lossless decoding speedups automatically.\n• Construct comprehensive benchmark suites across function calling, structured JSON extraction, coding, and multi-turn agentic reasoning to validate automated post-training recipes.\n• Partner with distributed training infrastructure teams to run serverless multi-node LoRA and full-weight RL jobs reliably on shared GPU pools.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, FSDP/Megatron, Verl/TRL, custom high-speed inference engine, Ray, ClickHouse, Kubernetes, H100/B200 GPU clusters. Workplace: Hybrid in Redwood City, CA. Compensation: $200,000 – $285,000 base salary plus generous unicorn equity, comprehensive medical/dental/vision, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 3+ years of machine learning engineering or research engineering experience focused on LLM fine-tuning, post-training alignment, model distillation, or compound AI routing.\n• Strong hands-on proficiency in PyTorch for training and fine-tuning dense and MoE transformers (LoRA/QLoRA, full-parameter SFT, DPO, GRPO/PPO).\n• Practical experience building automated LLM evaluation pipelines, LLM-as-a-judge calibrators, and synthetic training-data generators.\n• Solid statistical and optimization intuition for cost-quality-latency Pareto frontiers, calibration/uncertainty estimation, and contextual bandits.\n• Clean software engineering habits capable of turning experimental recipes into reliable, self-serve customer APIs.\n\nPREFERRED / BONUS EXPERIENCE\n• Published research or open-source work on LLM routing (RouteLLM/FrugalGPT-style cascades), speculative decoding draft-head training, or on-policy knowledge distillation.\n• Experience fine-tuning Vision-Language Models (VLMs) or multi-turn tool-calling agents.\n• Familiarity with serverless multi-tenant training job orchestration and memory-efficient optimizer states.",
      "updated_at": "2026-11-15T17:39:24.844Z",
      "created_at": "2026-11-15T16:50:05.984Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/machine-learning-engineer-model-routing-post",
        "markdown": "https://jobs.talent247.co/jobs/machine-learning-engineer-model-routing-post.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-model-routing-post/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-model-routing-post/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8cafae59-c035-45e7-9b05-6165a42110ba",
      "slug": "senior-llm-inference-engine-engineer-moe-lora",
      "title": "Senior LLM Inference Engine Engineer (MoE & LoRA)",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Redwood City, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 225000,
        "max": 325000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $325k",
        "subtitle": "annual package"
      },
      "summary": "Senior LLM Inference Engine Engineer (MoE & LoRA) (AI Infrastructure) — Redwood City, CA · Hybrid · $225k – $325k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar generative AI inference and post-training cloud headquartered in Redwood City, CA, founded by core creators of modern deep-learning frameworks. Serving trillions of daily tokens for top AI-native products and enterprises, their inference engine powers ultra-fast open-weight Mixture-of-Experts (MoE) models alongside thousands of concurrently hot-swapped custom LoRA and RL-tuned adapters on shared GPU pods.\n\nThey have retained Talent247 to appoint a Senior LLM Inference Engine Engineer (MoE & LoRA) in Redwood City, CA. Working at the core of their proprietary C++/CUDA/Python serving runtime, the successful candidate will push distributed MoE expert parallelism, disaggregated KV-caching, and multi-LoRA batching to industry-leading throughput-per-dollar frontiers.\n\nKEY RESPONSIBILITIES\n• Architect and optimize distributed Mixture-of-Experts (MoE) serving topologies—spanning wide expert parallelism (EP), RDMA/NVLink low-latency all-to-all token dispatch (DeepEP-class), and dynamic expert load balancing across multi-node Hopper and Blackwell pods.\n• Engineer zero-overhead multi-LoRA batched GEMM kernels and tiered GPU/CPU/NVMe adapter caches supporting thousands of customer fine-tuned models on shared base-weight replicas without cold-start stalls.\n• Implement disaggregated prefill/decode scheduling, speculative multi-token draft-verify loops (MTP / EAGLE-style), and grammar-constrained structured decoding with near-zero CPU overhead.\n• Optimize FP8, NVFP4, and mixed-precision quantization schemes preserving full reasoning and tool-calling fidelity on 100B–600B+ parameter architectures.\n• Deliver day-zero optimized support for newly released open-weight LLM, VLM, and audio-text architectures within hours of weight publication.\n• Profile end-to-end CUDA Graphs, NCCL/NVSHMEM collectives, and Python/C++ scheduler overhead using Nsight Compute and Nsight Systems.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: C++20, CUDA, CUTLASS, Triton, Python, PyTorch, custom SGLang/vLLM/TensorRT-LLM-class inference runtime, NVSHMEM/RDMA, H100/H200/B200 GPU clusters. Workplace: Hybrid in Redwood City, CA. Compensation: $225,000 – $325,000 base salary plus high-growth unicorn equity and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of deep learning runtime, GPU kernel, or LLM inference engine engineering experience in modern C++, CUDA, and Python.\n• Deep production familiarity with modern LLM serving internals (continuous batching, paged KV caches, chunked prefill, CUDA Graphs, tensor & expert parallelism).\n• Hands-on experience authoring or optimizing CUDA/CUTLASS/Triton kernels for attention, grouped GEMMs, MoE dispatch, or quantized matmuls (FP8/INT4/FP4).\n• Strong command of multi-GPU / multi-node interconnect performance (NVLink, InfiniBand/RoCE GPUDirect RDMA) and latency-throughput Pareto trade-offs.\n• Track record shipping rock-solid production runtime code handling billions of requests without memory leaks or numerical regressions.\n\nPREFERRED / BONUS EXPERIENCE\n• Core contributions to SGLang, vLLM, TensorRT-LLM, FlashInfer, DeepEP, Punica/S-LoRA, or PyTorch compiler/runtime internals.\n• Experience implementing multi-token prediction (MTP), EAGLE speculative decoding, or grammar-guided FSM token masking (`xgrammar`/`outlines`).\n• Familiarity with AMD Instinct (ROCm/HIP) or NVIDIA Blackwell GB200 NVL72 rack-scale optimization.",
      "updated_at": "2026-11-15T17:39:14.459Z",
      "created_at": "2026-11-15T16:50:02.166Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-llm-inference-engine-engineer-moe-lora",
        "markdown": "https://jobs.talent247.co/jobs/senior-llm-inference-engine-engineer-moe-lora.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-llm-inference-engine-engineer-moe-lora/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-llm-inference-engine-engineer-moe-lora/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "a7b2a7fd-943d-48a2-bc1a-ccbc4fac5296",
      "slug": "senior-ml-data-preference-pipeline-engineer",
      "title": "Senior ML Data & Preference Pipeline Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 205000,
        "max": 300000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$205k – $300k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Data & Preference Pipeline Engineer (MLOps & Platform) — San Francisco, CA · On-site · $205k – $300k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a category-defining AI-native software engineering company in San Francisco whose coding models and autonomous agents process billions of code completions and multi-file agentic trajectories every week. Turning live developer interaction patterns—strictly from opted-in telemetry and synthetic sandbox runs while guaranteeing airtight zero-data-retention isolation for enterprise tenants—into clean, real-time post-training preference signals is a cornerstone of their data flywheel.\n\nThey have engaged Talent247 to hire a Senior ML Data & Preference Pipeline Engineer onsite in San Francisco, CA. Working shoulder-to-shoulder with code-model researchers, the appointed engineer will build ultra-high-throughput streaming ingestion, privacy-enforcing sanitization, and automated preference-pair synthesis engines feeding daily SFT and RL runs.\n\nKEY RESPONSIBILITIES\n• Architect high-throughput Rust, Go, and Python streaming pipelines ingesting tens of gigabytes per second of structured editor/agent telemetry into ClickHouse, Kafka, and petabyte columnar lakes.\n• Build deterministic, provably audited privacy and tenant-isolation gates ensuring enterprise zero-data-retention traffic, secrets/API keys, and PII never enter persistence or training pipelines.\n• Design counterfactual preference-extraction pipelines reconstructing exact repository states, accepted vs. modified multi-line diffs, compiler/linter outcomes, and long-horizon survival of agent-authored code.\n• Engineer fast AST-aware deduplication, near-duplicate MinHash clustering, syntax/test-validity filtering, and dynamic curriculum-mixing pipelines feeding multi-node PyTorch training clusters.\n• Construct real-time online experiment attribution dashboards linking training-mixture ablations directly to sub-hour live user acceptance, edit-survival, and latency metrics.\n• Optimize petabyte storage layouts (Lance / Parquet / Arrow) and zero-copy tokenized shard streaming to keep thousand-GPU training runs 100% compute-bound.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust, Python, ClickHouse, Kafka / Redpanda, Ray Data, Lance / Apache Arrow / Parquet, tree-sitter, PyTorch data loaders, Kubernetes, AWS/GCP. Workplace: Onsite in San Francisco, CA. Compensation: $205,000 – $300,000 base salary plus top-tier hyper-growth equity, comprehensive benefits, and daily meals.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of data infrastructure, ML data platform, or high-throughput streaming systems engineering experience operating at multi-terabyte-per-day scale.\n• Strong systems programming fluency in Rust, C++, or Go paired with idiomatic Python for ML dataset tooling and PyTorch integration.\n• Production mastery of analytical/streaming stores such as ClickHouse, Kafka/Redpanda, Ray Data, and modern columnar formats (Parquet, Arrow, Lance).\n• Proven experience building ML training data engines—covering deduplication, quality filtering, secret/PII scrubbing, or RLHF preference-pair construction.\n• Extreme rigor around data correctness, privacy boundary enforcement, and reproducible dataset lineage.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience parsing and transforming source-code corpora at scale using `tree-sitter`, compiler ASTs, or git commit-graph mining.\n• Background building online feature/bandit feedback loops or real-time A/B experimentation platforms.\n• Familiarity with LLM tokenization internals, sequence packing, and deterministic multi-epoch data mixing.",
      "updated_at": "2026-11-15T17:39:04.630Z",
      "created_at": "2026-11-15T16:49:57.801Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-data-preference-pipeline-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-data-preference-pipeline-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-data-preference-pipeline-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-data-preference-pipeline-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "0e2ecffd-097e-4af1-b6f5-0a8f20a7cded",
      "slug": "staff-agentic-runtime-sandbox-engineer",
      "title": "Staff Agentic Runtime & Sandbox Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Palo Alto, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 235000,
        "max": 340000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$235k – $340k",
        "subtitle": "annual package"
      },
      "summary": "Staff Agentic Runtime & Sandbox Engineer (Applied AI) — Palo Alto, CA · On-site · $235k – $340k. THE CLIENT & MANDATE Our client is an iconic AI-native…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an iconic AI-native developer platform scaling across San Francisco and Palo Alto, CA, evolving software engineering from interactive in-editor assistance to fleets of autonomous background coding agents that clone repositories, spin up full microservice stacks, verify UI changes in headless browsers, and submit review-ready pull requests in parallel.\n\nThey have partnered with Talent247 to appoint a Staff Agentic Runtime & Sandbox Engineer onsite in Palo Alto, CA. Architecting at the intersection of operating-system virtualization, distributed storage, and language-server tooling, the successful candidate will build the instant-clone microVM cloud where millions of autonomous coding agents execute safely every day.\n\nKEY RESPONSIBILITIES\n• Architect sub-second Copy-on-Write (CoW) Firecracker/KVM microVM snapshotting, memory restore, and content-addressable filesystem layers booting warm multi-gigabyte monorepo environments in <250ms.\n• Build high-speed, low-overhead agent tool daemons in Rust and TypeScript exposing incremental LSP diagnostics, tree-sitter AST queries, fast ripgrep/vector indices, headless browser inspection, and PTY terminal multiplexing.\n• Design hermetic network egress proxies, secret-injection brokers, and fine-grained syscall/filesystem isolation so untrusted agent-generated code can run Docker Compose, databases, and unit suites safely inside multi-tenant infrastructure.\n• Engineer stateful session hibernation, live workspace syncing between cloud sandboxes and local developer machines, and deterministic trajectory replay for RL post-training.\n• Scale regional bare-metal and cloud hypervisor fleets handling extreme diurnal burstiness across hundreds of thousands of concurrent background agent runs.\n• Partner directly with RL and agentic model researchers so the same high-speed sandbox runtime powers both live user agents and million-rollout RL training loops.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Rust, TypeScript, Go, Linux KVM / Firecracker / Cloud Hypervisor, overlayfs / FUSE / ZFS / NVMe-oF, eBPF/nftables, LSP / tree-sitter, Kubernetes & bare-metal Linux hosts. Workplace: Onsite in Palo Alto, CA. Compensation: $235,000 – $340,000 base salary plus high-upside hyper-growth equity, comprehensive healthcare, daily meals, and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 6+ years of systems infrastructure engineering experience building cloud sandboxes, serverless container/VM runtimes, distributed filesystems, or developer cloud environments.\n• Deep production mastery of Rust (or modern C++/Go) and Linux kernel primitives: KVM/Firecracker virtualization, namespaces/cgroups, memory snapshotting (`userfaultfd`), overlay/CoW filesystems, and virtual networking.\n• Strong track record eliminating cold-start latency, tail-latency IO amplification, and noisy-neighbor contention on dense multi-tenant bare-metal hosts.\n• Solid security engineering grounding in hypervisor isolation, egress filtering, and credential brokering.\n• High-agency builder mentality excited to ship foundational infrastructure in small, autonomous engineering squads.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience building cloud IDEs, CI/CD execution engines, serverless GPU/CPU microVM platforms, or RL coding-environment clusters.\n• Familiarity with Language Server Protocol (LSP) internals, Nix/Bazel remote caching, or git virtual filesystems for massive monorepos.\n• Open-source contributions to Firecracker, Cloud Hypervisor, containerd, Tokio, or developer tooling ecosystems.",
      "updated_at": "2026-11-15T17:38:56.118Z",
      "created_at": "2026-11-15T16:49:54.483Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-agentic-runtime-sandbox-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-agentic-runtime-sandbox-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-agentic-runtime-sandbox-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-agentic-runtime-sandbox-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "19fe8089-228e-4391-bd00-2fd1bb280f81",
      "slug": "senior-machine-learning-engineer-low-latency-code",
      "title": "Senior Machine Learning Engineer, Low-Latency Code Models",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 220000,
        "max": 330000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$220k – $330k",
        "subtitle": "annual package"
      },
      "summary": "Senior Machine Learning Engineer, Low-Latency Code Models (Machine Learning) — San Francisco, CA · On-site · $220k – $330k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is the fastest-growing AI-native software engineering platform in history, headquartered in San Francisco and trusted daily by millions of professional developers and majority-adoption engineering teams across the Fortune 500. Unlike generic chat models, their competitive edge comes from purpose-built, ultra-low-latency code models that predict multi-line edits, cursor jumps, and cross-file refactors faster than human thought (<100ms end-to-end).\n\nThey have retained Talent247 to hire a Senior Machine Learning Engineer for Low-Latency Code Models onsite in San Francisco, CA. Joining a high-density, low-ego research and engineering team, the appointed engineer will co-design custom speculative architectures, online preference RL loops, and repository-context rankers powering billions of daily code completions.\n\nKEY RESPONSIBILITIES\n• Train and iterate on specialized dense and sparse Mixture-of-Experts (MoE) code models optimized for multi-hunk next-edit prediction, diff application, and cross-file semantic navigation.\n• Pioneer speculative edit-decoding algorithms—exploiting the high overlap between existing source files and proposed edits—to deliver 1,000+ tok/s effective edit speeds under strict sub-100ms p95 SLAs.\n• Co-design repository-wide context retrieval and AST/LSP-aware token compression models that pack exact type definitions, call-graph neighbors, and recent edit history into ultra-compact prompts.\n• Run continuous online distillation and reinforcement learning (DPO/GRPO) from real-world implicit edit acceptance/rejection signals while strictly honoring enterprise zero-data-retention boundaries.\n• Partner side-by-side with custom inference kernel engineers to co-optimize model depth/width, KV-cache reuse across keystrokes, and FP8/FP4 quantization.\n• Own live A/B experimentation and offline replay harnesses measuring keystroke savings, edit precision, and zero-distraction calibration.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, C++/CUDA/Triton, Megatron/FSDP, SGLang/TensorRT-LLM/custom speculative serving engines, tree-sitter/LSP, multi-thousand H100/B200 GPU clusters. Workplace: Onsite in San Francisco, CA. Compensation: $220,000 – $330,000 base salary plus premier hyper-growth equity, daily catered meals, and full medical/401(k) benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of machine learning research engineering experience training or post-training large transformer models, code models, or ultra-low-latency neural systems.\n• Deep hands-on mastery of PyTorch distributed training, model distillation, SFT/RLHF/GRPO post-training, and sequence-level loss engineering.\n• Strong systems intuition for transformer inference bottlenecks—KV-cache prefix sharing, speculative draft-verify decoding, batching dynamics, and quantization trade-offs.\n• Exceptional software engineering taste: comfortable jumping from a PyTorch training loop into Rust/TypeScript editor context assembly or CUDA profiling.\n• Passion for developer tools and building AI models whose quality you feel personally in your own workflow every hour.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior work on code generation models, speculative/medusa/grammatical decoding, diff-representation tokenization, or neural code search.\n• Familiarity with compiler/IDE internals (LSP, tree-sitter, control-flow/type graphs) as structured conditioning signals for LLMs.\n• Publications at NeurIPS, ICML, ICLR, or PLDI/ICSE, or standout open-source ML/devtools contributions.",
      "updated_at": "2026-11-15T17:38:46.893Z",
      "created_at": "2026-11-15T16:49:51.400Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-low-latency-code",
        "markdown": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-low-latency-code.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-low-latency-code/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-low-latency-code/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "9258ec89-9a3c-48b8-b513-0e5e5c9b2dd5",
      "slug": "staff-ml-platform-cluster-reliability-engineer",
      "title": "Staff ML Platform & Cluster Reliability Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 260000,
        "max": 350000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$260k – $350k",
        "subtitle": "annual package"
      },
      "summary": "Staff ML Platform & Cluster Reliability Engineer (MLOps & Platform) — San Francisco, CA · Hybrid · $260k – $350k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco frontier AI laboratory operating a planet-scale heterogeneous compute fleet spanning hundreds of thousands of GPUs and custom AI accelerators across multiple cloud providers and bare-metal datacenters. Keeping training superclusters and global low-latency inference regions operating as a single, self-healing compute fabric is essential to both research velocity and commercial API reliability.\n\nThey have partnered with Talent247 to appoint a Staff ML Platform & Cluster Reliability Engineer in San Francisco, CA. Acting as a technical anchor across cluster orchestration, fleet health automation, and storage streaming, the successful candidate will build the Go and Rust control planes that maximize goodput across billions of dollars of AI compute.\n\nKEY RESPONSIBILITIES\n• Architect topology-aware Kubernetes gang schedulers and multi-cluster workload placement engines optimizing NVLink/InfiniBand locality across pre-training, RL rollout, and global inference fleets.\n• Build autonomous hardware validation, continuous active/passive health probing, and automated node-quarantine/drain pipelines eliminating GPU Xid faults, flapping optical links, and silent compute corruption without human paging.\n• Design high-throughput petabyte checkpoint and model-weight distribution fabrics in Rust and Go capable of saturating 400G/800G NICs from object storage into GPU memory in seconds.\n• Establish fleet-wide goodput accounting, preemption-aware job elasticity, and unified eBPF/OpenTelemetry observability correlating kernel/driver telemetry with job-level step times.\n• Lead multi-region disaster-recovery exercises, capacity ramp-ups for newly delivered 10,000+ accelerator clusters, and blameless architectural post-mortems.\n• Mentor senior infrastructure engineers while partnering closely with distributed training leads and cloud/hardware vendor engineering teams.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Go, Rust, Python, Kubernetes (Kueue / custom operators & CNI/DRA plugins), Linux kernel/eBPF, InfiniBand/RoCEv2, Prometheus/ClickHouse, multi-cloud & bare-metal GPU/ASIC clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $260,000 – $350,000 base salary plus flagship frontier-lab equity and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of infrastructure, distributed systems, or production SRE/platform engineering experience, including 3+ years operating large-scale GPU or AI accelerator clusters.\n• Expert-level systems development in Go or Rust alongside deep mastery of Kubernetes internals (custom schedulers, CRDs/operators, Dynamic Resource Allocation, device plugins).\n• Deep Linux systems troubleshooting expertise across PCIe/NVLink topologies, GPU drivers/DCGM, RDMA/RoCEv2/InfiniBand networking, and kernel memory/IO subsystems.\n• Proven track record improving cluster utilization/goodput and building automated self-healing remediation at multi-thousand-node scale.\n• Calm technical leadership across high-stakes cluster bring-ups and production reliability initiatives.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience building P2P or RDMA-accelerated container/model-weight distribution systems across thousands of GPU nodes.\n• Familiarity with both synchronous pre-training gang scheduling and elastic asynchronous RL actor-learner cluster workloads.\n• Core contributions to Kubernetes WG-Serving / Kueue, Volcano, Ray on K8s, or Linux eBPF observability tools.",
      "updated_at": "2026-11-15T17:38:36.784Z",
      "created_at": "2026-11-15T16:49:47.908Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-ml-platform-cluster-reliability-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-ml-platform-cluster-reliability-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-ml-platform-cluster-reliability-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-ml-platform-cluster-reliability-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "083ad786-2cf7-40a4-8599-38146d6995fe",
      "slug": "senior-ml-compiler-engineer-long-context-inference",
      "title": "Senior ML Compiler Engineer, Long-Context Inference",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Seattle, WA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 245000,
        "max": 340000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$245k – $340k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Compiler Engineer, Long-Context Inference (AI Infrastructure) — Seattle, WA · Hybrid · $245k – $340k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a frontier AI research and deployment company operating a multi-accelerator compute fleet across NVIDIA GPUs and custom cloud AI ASICs/TPUs, with a dedicated compiler and performance engineering hub in Seattle, WA. Serving flagship models with 200k–1M+ token context windows across multi-hour coding and research sessions places extreme pressure on compiler graph lowering, HBM memory scheduling, and cross-chip collective fusion.\n\nThey have retained Talent247 to hire a Senior ML Compiler Engineer for Long-Context Inference in Seattle, WA. The successful candidate will build multi-target XLA, MLIR, and custom kernel-generation passes that maximize throughput and slash time-to-first-token across heterogeneous frontier accelerator fleets.\n\nKEY RESPONSIBILITIES\n• Design and implement XLA, MLIR, and StableHLO compiler passes optimizing ultra-long-context (200k–1M+ token) prefill and decode graphs across GPUs and custom systolic-array AI ASICs/TPUs.\n• Architect compiler-driven compute-communication overlap transformations (async collective matmuls, Ring/Ulysses context-parallel attention fusion, and pipelined all-to-all MoE dispatch).\n• Optimize HBM-to-SRAM/shared-memory tiling, KV-cache quantization (FP8/INT4), prefix-tree paging layouts, and rematerialization schedules to fit massive batch sizes within tight memory envelopes.\n• Author custom Pallas, Triton, and CUDA/C++ escape-hatch kernels where general compiler lowering leaves roof-line performance on the table, then upstream clean lowering patterns back into the compiler.\n• Build automated cross-backend numerical parity checkers and compiler-regression performance gates across dozens of active production model variants.\n• Partner with model architecture scientists on hardware-co-designed attention mechanisms (MLA, GQA, hybrid sliding-window/global attention) optimized for multi-accelerator compilation.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: C++20, MLIR, OpenXLA / StableHLO, LLVM, Python, JAX, PyTorch/TorchXLA, Pallas, Triton, CUDA, multi-pod GPU & cloud TPU/ASIC clusters. Workplace: Hybrid in Seattle, WA. Compensation: $245,000 – $340,000 base salary plus substantial frontier-lab equity, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of ML compiler, GPU/ASIC toolchain, or high-performance deep learning runtime engineering experience in modern C++ and Python.\n• Deep internals knowledge of OpenXLA (HLO passes, layout assignment, memory space scheduling), MLIR, TVM, or TorchInductor.\n• Strong command of accelerator hardware bottlenecks (systolic arrays, tensor cores, HBM bandwidth, ICI/NVLink interconnect topologies) and roof-line analysis.\n• Practical familiarity with modern long-context LLM serving mechanics—paged KV caches, chunked prefill, context parallelism, and speculative decoding.\n• Proven ability to debug subtle compiler miscompilations, NaN/precision divergence, and multi-host collective deadlocks.\n\nPREFERRED / BONUS EXPERIENCE\n• Hands-on experience writing Pallas (TPU/GPU) or Triton/CUTLASS kernels for custom attention or quantized matrix multiplication.\n• Experience targeting multiple distinct AI hardware backends (e.g., NVIDIA Hopper/Blackwell + Google Cloud TPU or AWS Trainium) from a unified model codebase.\n• Upstream commit history in OpenXLA, MLIR, LLVM, JAX, or Triton.",
      "updated_at": "2026-11-15T17:38:26.903Z",
      "created_at": "2026-11-15T16:49:43.698Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-compiler-engineer-long-context-inference",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-compiler-engineer-long-context-inference.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-compiler-engineer-long-context-inference/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-compiler-engineer-long-context-inference/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "90950331-2bef-4298-b424-ed8966f36b24",
      "slug": "applied-ai-security-guardrails-architect",
      "title": "Applied AI Security & Guardrails Architect",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 255000,
        "max": 345000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$255k – $345k",
        "subtitle": "annual package"
      },
      "summary": "Applied AI Security & Guardrails Architect (Applied AI) — New York, NY · Hybrid · $255k – $345k. THE CLIENT & MANDATE Our client is a leading…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a leading safety-focused frontier AI company headquartered in San Francisco with a rapidly expanding enterprise and security engineering presence in New York City. As global banks, sovereign institutions, healthcare networks, and software platforms connect autonomous coding and knowledge agents directly to internal production infrastructure via Model Context Protocol (MCP) and computer-use runtimes, defense-in-depth agent security has become paramount.\n\nThey have partnered with Talent247 to appoint an Applied AI Security & Guardrails Architect in New York, NY. Sitting across security engineering, applied AI, and strategic enterprise architecture, the successful candidate will design zero-trust agent sandboxes, indirect-prompt-injection defenses, and cryptographic capability-control frameworks protecting the world's most security-sensitive AI deployments.\n\nKEY RESPONSIBILITIES\n• Architect defense-in-depth runtime security patterns for tool-using and computer-use agents—spanning capability-scoped MCP gateways, microVM/network egress sandboxing, and deterministic data-exfiltration controls.\n• Design layered ML + deterministic guardrail pipelines mitigating indirect prompt injection across untrusted web pages, emails, code repositories, and third-party SaaS records.\n• Partner directly with CISO and Principal Security Architects at flagship financial services, life sciences, and technology customers to threat-model multi-agent production rollouts.\n• Build automated continuous red-teaming harnesses that simulate confused-deputy attacks, credential harvesting, MCP tool poisoning, and cross-tenant memory leakage.\n• Co-design enterprise governance primitives with core platform teams—including cryptographic audit trails, zero-data-retention enclaves, customer-managed KMS keys, and policy-as-code tool approvals.\n• Publish authoritative reference architectures and open security standards for safe enterprise agent deployment.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust, TypeScript, Go, Model Context Protocol (MCP), Firecracker/gVisor/seccomp-bpf, Envoy/mTLS, OAuth 2.1 / OIDC, cloud KMS/confidential computing, frontier LLM classifier guardrails. Workplace: Hybrid in New York, NY. Compensation: $255,000 – $345,000 base salary plus high-value frontier lab equity and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of security architecture, application/cloud security engineering, or secure distributed systems development, including 2+ years focused on LLM/agent security.\n• Deep technical understanding of AI-native attack surfaces: indirect prompt injection, tool/MCP schema poisoning, sandbox escape, data exfiltration side-channels, and jailbreak chaining.\n• Strong systems and coding fluency (Python, Go, Rust, or TypeScript) capable of building working reference sandboxes, policy proxies, and adversarial exploit harnesses.\n• Mastery of enterprise cloud security controls (AWS/GCP/Azure IAM, VPC isolation, OAuth2/OIDC token scoping, zero-trust networking, confidential computing).\n• Clear, pragmatic communication style trusted equally by internal alignment researchers and Fortune 100 CISOs.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior offensive security research, bug-bounty discoveries, or published research on securing tool-using AI agents.\n• Experience securing developer IDE/CLI coding agents or enterprise SaaS connector platforms at scale.\n• Contributions to OWASP GenAI, MCP security specifications, or open-source sandboxing projects.",
      "updated_at": "2026-11-15T17:38:16.239Z",
      "created_at": "2026-11-15T16:49:39.588Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/applied-ai-security-guardrails-architect",
        "markdown": "https://jobs.talent247.co/jobs/applied-ai-security-guardrails-architect.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/applied-ai-security-guardrails-architect/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/applied-ai-security-guardrails-architect/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "0e92c376-facb-4394-9ca1-9299591a2af4",
      "slug": "research-engineer-agentic-coding-environments",
      "title": "Research Engineer, Agentic Coding Environments",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 260000,
        "max": 360000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$260k – $360k",
        "subtitle": "annual package"
      },
      "summary": "Research Engineer, Agentic Coding Environments (Machine Learning) — San Francisco, CA · Hybrid · $260k – $360k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco frontier AI laboratory whose flagship models set the industry benchmark for autonomous software engineering, terminal orchestration, and long-horizon computer use. Behind those coding breakthroughs sits an elite Agentic Coding Environments research engineering group building the realistic, multi-hour software sandboxes and reinforcement learning curricula where frontier models learn to navigate, architect, debug, and verify production-grade codebases.\n\nThey have engaged Talent247 to hire a Research Engineer for Agentic Coding Environments in San Francisco, CA. Bridging RL research and systems engineering, the successful candidate will build massively scalable repository gyms, compiler/execution verifiers, and multi-turn post-training loops pushing autonomous coding capabilities from single-PR patches to complex multi-service engineering projects.\n\nKEY RESPONSIBILITIES\n• Architect hermetic, sub-second containerized and microVM coding environments supporting polyglot build systems (Rust, C++, Go, TypeScript, Python, Java, Kubernetes, databases) across tens of thousands of parallel RL rollouts.\n• Design rich, cheat-proof verification harnesses combining hidden integration tests, property-based fuzzing, performance/memory regression benchmarks, and calibrated code-review reward models.\n• Run large-scale multi-turn reinforcement learning experiments teaching models long-horizon repository exploration, hypothesis-driven debugging, git history archaeology, and clean architectural refactoring.\n• Build automated task-generation pipelines that synthesize realistic feature migrations, subtle concurrency bugs, and security patches across complex multi-repo topologies without benchmark contamination.\n• Analyze long-context (100k+ token) agent trajectories to improve context-compaction habits, shell/editor tool ergonomics, and self-correction when builds fail.\n• Partner closely with developer-product and safety teams to ensure autonomous coding agents respect sandbox boundaries and avoid destructive operational actions.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust, TypeScript, PyTorch/JAX, Firecracker/Docker/Kubernetes rollout clusters, LSP/tree-sitter/Bazel/Nix build toolchains, frontier GPU/ASIC training clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $260,000 – $360,000 base salary plus substantial equity grants, full healthcare, and generous benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of software systems and ML/RL research engineering experience with exceptional polyglot coding craftsmanship.\n• Hands-on experience building coding-agent harnesses, SWE-bench-style containerized evaluation/RL gyms, or large-scale LLM post-training pipelines.\n• Deep systems fluency in Linux containerization, microVM snapshotting, reproducible polyglot builds (Docker, Nix, Bazel), and distributed task scheduling.\n• Strong intuition for how language models interact with shell terminals, file editors, debuggers, and test runners—and how they attempt to game naive graders.\n• Ability to own both the systems reliability of 100k-container rollout fleets and the scientific analysis of RL training dynamics.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience training frontier or open-weight coding models with online RL (GRPO/PPO) over multi-turn interactive environments.\n• Background in compilers, static analysis, fuzzing, OS internals, or developer tools engineering.\n• Track record of high-impact open-source contributions or strong competitive programming achievements.",
      "updated_at": "2026-11-15T17:38:06.207Z",
      "created_at": "2026-11-15T16:49:35.885Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-engineer-agentic-coding-environments",
        "markdown": "https://jobs.talent247.co/jobs/research-engineer-agentic-coding-environments.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-engineer-agentic-coding-environments/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-engineer-agentic-coding-environments/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3316318b-1e61-4cc4-9761-691dcb9477f9",
      "slug": "research-scientist-alignment-mechanistic",
      "title": "Research Scientist, Alignment & Mechanistic Interpretability",
      "department": "AI Research",
      "department_code": "RES",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 280000,
        "max": 380000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$280k – $380k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Alignment & Mechanistic Interpretability (AI Research) — San Francisco, CA · Hybrid · $280k – $380k. THE CLIENT & MANDATE Our client…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a mission-driven frontier AI research laboratory headquartered in San Francisco, renowned globally for pioneering constitutional alignment, mechanistic interpretability, and frontier coding and reasoning models. Their Interpretability & Alignment Science organization seeks to reverse-engineer the internal computational circuits of multi-hundred-billion-parameter transformers so safety guarantees rest on mechanistic understanding rather than black-box behavioral spot checks alone.\n\nThey have retained Talent247 to appoint a Research Scientist in Alignment & Mechanistic Interpretability in San Francisco, CA. Working alongside pioneers of dictionary learning and attribution graphs, the appointed scientist will scale interpretability decompositions to flagship production models and translate circuit-level discoveries into concrete training-time and inference-time alignment interventions.\n\nKEY RESPONSIBILITIES\n• Originate and scale dictionary-learning architectures (sparse autoencoders, transcoders, cross-layer attribution graphs) across 70B+ to frontier-scale multimodal and reasoning transformers.\n• Trace internal computational circuits governing multi-step planning, chain-of-thought faithfulness, sycophancy, sandbagging, reward hacking, and refusal behavior.\n• Design causal activation-patching, feature-steering, and representation-monitoring interventions that detect deceptive or misaligned latent states during online RL and real-world agent deployment.\n• Partner with distributed systems engineers to build petabyte-activation caching and low-overhead auxiliary dictionary-training pipelines integrated directly into flagship training clusters.\n• Construct rigorous quantitative benchmarks evaluating whether interpretability tools genuinely improve human/automated auditing of subtle sleeper-agent or specification-gaming model organisms.\n• Publish foundational open research advancing the broader AI safety and mechanistic interpretability community.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, JAX, Triton custom sparse-dictionary kernels, distributed activation-streaming pipelines, multi-thousand GPU/ASIC research clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $280,000 – $380,000 base salary plus generous frontier-lab equity, comprehensive health & wellness benefits, and relocation/visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Physics, Neuroscience, Mathematics, Computer Science, or equivalent frontier research track record in interpretability, alignment, or deep learning science.\n• Demonstrated originality through influential papers, technical reports, or open-source work in mechanistic interpretability, representation engineering, scalable oversight, or alignment stress-testing.\n• Strong hands-on fluency in PyTorch or JAX—comfortable writing clean distributed experiments over large transformer internals (residual streams, attention heads, MLP activations).\n• Deep scientific skepticism: habit of testing competing null hypotheses and designing falsifiable causal ablations rather than relying on cherry-picked qualitative examples.\n• Collaborative communication style suited to cross-pollinating fundamental alignment science with production post-training teams.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience training sparse autoencoders (SAEs), cross-layer transcoders, or circuit-attribution pipelines on large language models.\n• Experience constructing alignment \"model organisms\" to stress-test auditing, unlearning, or RLHF robustness techniques.\n• Familiarity with custom Triton/CUDA sparse matrix kernels for scaling million-feature dictionary learning.",
      "updated_at": "2026-11-15T17:37:56.522Z",
      "created_at": "2026-11-15T16:49:32.061Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-alignment-mechanistic",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-alignment-mechanistic.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-alignment-mechanistic/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-alignment-mechanistic/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "24938f34-62b9-4b13-b695-bf754a3603a3",
      "slug": "senior-machine-learning-engineer-multimodal",
      "title": "Senior Machine Learning Engineer, Multimodal Evaluations",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 235000,
        "max": 320000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$235k – $320k",
        "subtitle": "annual package"
      },
      "summary": "Senior Machine Learning Engineer, Multimodal Evaluations (Machine Learning) — New York, NY · Hybrid · $235k – $320k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a flagship frontier AI laboratory expanding their East Coast research and engineering hub in New York City. As frontier models unify native speech-to-speech audio, high-resolution vision, video understanding, and autonomous computer control, static text benchmarks fail to capture real-world capability or failure modes—making next-generation multimodal evaluation systems essential to every model training decision.\n\nThey have retained Talent247 to hire a Senior Machine Learning Engineer for Multimodal Evaluations in New York, NY. Working directly alongside post-training researchers and product launch leads, the appointed engineer will build interactive, contamination-free evaluation environments measuring real-time voice, visual reasoning, UI navigation, and cross-modal tool use.\n\nKEY RESPONSIBILITIES\n• Architect scalable, low-variance multimodal evaluation pipelines testing native audio prosody/interruption handling, dense chart/diagram spatial reasoning, temporal video grounding, and GUI computer use.\n• Build containerized OS, browser, and code-execution sandboxes where multimodal agents are graded deterministically on end-state environment mutations rather than brittle string matching.\n• Train and calibrate specialized multimodal judge models and rubric verifiers validated rigorously against expert human preference panels.\n• Design automated data-contamination scanners, dynamic benchmark refresh pipelines, and psychometric difficulty-scaling analyses preventing Goodhart overfitting during post-training.\n• Own automated pre-merge and pre-release regression gates running across thousands of checkpoints during active frontier training campaigns.\n• Partner with NYC and SF research teams to turn fine-grained error taxonomies directly into targeted post-training data mixtures.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Rust/TypeScript, containerized VM/browser sandboxes, real-time WebRTC/audio eval harnesses, vLLM/custom inference clusters, Kubernetes. Workplace: Hybrid in New York, NY. Compensation: $235,000 – $320,000 base salary plus top-tier frontier lab equity, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of machine learning engineering or research engineering experience working on multimodal foundation models, speech/vision systems, or large-scale LLM evaluation infrastructure.\n• Strong hands-on proficiency in PyTorch, distributed batch/streaming inference orchestration, and statistical benchmark design (confidence intervals, power analysis, Elo/Bradley-Terry modeling).\n• Proven ability to architect reliable sandboxed execution harnesses (Docker, Playwright/OS VMs, audio streaming simulators) at high concurrency.\n• Experience diagnosing subtle failure modes across vision encoders, audio tokenizers, OCR/grounding coordinates, or multi-turn agent trajectories.\n• Crisp engineering discipline writing maintainable, reproducible evaluation frameworks trusted for flagship launch decisions.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior authorship of widely cited multimodal, coding, or computer-use benchmarks or evaluation papers at CVPR, NeurIPS, ICML, ICLR, or Interspeech.\n• Experience with full-duplex speech-to-speech evaluation metrics (turn-taking latency, acoustic emotion faithfulness, barge-in robustness).\n• Familiarity with RLVR grader hardening against reward hacking during online RL.",
      "updated_at": "2026-11-15T17:37:36.759Z",
      "created_at": "2026-11-15T16:49:25.257Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-multimodal",
        "markdown": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-multimodal.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-multimodal/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-multimodal/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "29cc7e06-7aa9-4f88-a963-31da227d1782",
      "slug": "senior-applied-ai-engineer-enterprise-agents",
      "title": "Senior Applied AI Engineer, Enterprise Agents",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 225000,
        "max": 310000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $310k",
        "subtitle": "annual package"
      },
      "summary": "Senior Applied AI Engineer, Enterprise Agents (Applied AI) — San Francisco, CA · Hybrid · $225k – $310k. THE CLIENT & MANDATE Our client is a globally…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally recognized frontier AI research and product organization in San Francisco whose reasoning, multimodal, and coding APIs power mission-critical workflows across the Fortune 500 and global AI-native startups. Their Applied Enterprise Agents engineering team embeds directly with flagship strategic partners to architect autonomous, multi-agent systems that redefine software development, financial analysis, customer operations, and scientific R&D.\n\nThey have engaged Talent247 to recruit a Senior Applied AI Engineer for Enterprise Agents in San Francisco, CA. Combining production backend craftsmanship with deep post-training and evaluation intuition, the successful candidate will ship resilient multi-agent architectures, custom reinforcement fine-tuning loops, and low-latency voice/text agent harnesses into high-volume production.\n\nKEY RESPONSIBILITIES\n• Architect production multi-agent orchestration graphs featuring typed tool-calling contracts, context compaction, parallel sub-agent delegation, and deterministic safety guardrails.\n• Build automated task-specific evaluation harnesses, trajectory graders, and synthetic adversarial test suites driving measurable accuracy gains across complex multi-turn workflows.\n• Lead reinforcement fine-tuning (RFT), supervised distillation, and prompt-caching optimizations to meet strict sub-second p95 latency and unit-economic targets at millions of daily requests.\n• Co-engineer reference Model Context Protocol (MCP) connectors, sandboxed code-execution environments, and enterprise RBAC/audit-logging integrations alongside customer Staff engineers.\n• Synthesize recurring developer friction points and capability gaps directly into product specs for core API, tool-use post-training, and agent SDK teams.\n• Author authoritative technical blueprints and reusable evaluation libraries adopted across global enterprise engineering deployments.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, TypeScript, Go, frontier reasoning & multimodal voice/vision APIs, reinforcement fine-tuning pipelines, MCP servers, Docker/microVM sandboxes, OpenTelemetry, Kubernetes. Workplace: Hybrid in San Francisco, CA. Compensation: $225,000 – $310,000 base salary plus substantial frontier-lab equity and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of software engineering and applied machine learning experience building and scaling production distributed applications and LLM-powered agent systems.\n• Deep practical mastery of agentic system design: multi-turn tool orchestration, structured outputs, hybrid retrieval, trajectory-level evaluation, and latency optimization.\n• Hands-on experience fine-tuning language models (SFT, LoRA, DPO, or reinforcement fine-tuning) and building calibrated automated graders.\n• Strong full-stack/backend fluency in Python and TypeScript (plus Go or Rust advantageous) with clean systems architecture habits.\n• Track record operating effectively with senior technical leaders at complex enterprise or hyper-growth technology organizations.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience shipping real-time duplex voice agents, browser/computer-use agents, or autonomous software-engineering coding agents.\n• Strong familiarity with enterprise cloud security (SOC2, VPC peering, zero-data-retention architectures, OAuth2/OIDC delegation).\n• Open-source agent SDK, evaluation framework, or MCP ecosystem contributions.",
      "updated_at": "2026-11-15T17:37:27.677Z",
      "created_at": "2026-11-15T16:49:21.026Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-enterprise-agents",
        "markdown": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-enterprise-agents.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-enterprise-agents/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-enterprise-agents/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "e72b3c09-7961-4924-81dc-2581a3022687",
      "slug": "staff-distributed-training-systems-engineer",
      "title": "Staff Distributed Training Systems Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 290000,
        "max": 385000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$290k – $385k",
        "subtitle": "annual package"
      },
      "summary": "Staff Distributed Training Systems Engineer (AI Infrastructure) — San Francisco, CA · On-site · $290k – $385k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco frontier AI laboratory operating some of the largest single-job GPU training supercomputers ever assembled. At tens of thousands of interconnected accelerators, hardware failures, silent data corruption, and network congestion become continuous statistical certainties rather than rare anomalies—making distributed training systems engineering a decisive moat for frontier model scaling.\n\nThey have partnered with Talent247 to hire a Staff Distributed Training Systems Engineer onsite in San Francisco, CA. Serving as a technical lead within their Compute & Training Foundations pillar, the appointed engineer will architect elastic fault-tolerant runtime primitives, collective communication schedules, and zero-stall checkpointing systems sustaining >95% effective cluster goodput across multi-month pre-training and RL runs.\n\nKEY RESPONSIBILITIES\n• Architect 5D-parallel distributed training runtimes (data, tensor, pipeline, context, and expert parallelism) scaling cleanly across 10,000 to 100,000+ next-generation GPUs.\n• Design sub-minute fault detection, dynamic rank reconfiguration, and in-memory/NVMe asynchronous checkpointing engines that recover from GPU or link failures without full-job restarts.\n• Co-optimize NCCL collective communication schedules, computation-communication overlap kernels, and topology-aware job placement over 400G/800G InfiniBand and RoCEv2 fabrics.\n• Build deterministic silent-data-corruption (SDC) detectors, deterministic replay profilers, and per-iteration straggler attribution pipelines catching degraded silicon in flight.\n• Collaborate directly with model architecture researchers to make novel sparse MoE routing, ultra-long-context attention, and FP8/FP4 low-precision training recipes hardware-optimal from day one.\n• Mentor senior distributed systems and GPU kernel engineers across training runtime, cluster scheduling, and networking squads.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: C++20, CUDA, Python, PyTorch internals (c10d / TorchDistributed), NCCL, InfiniBand/RoCEv2 GPUDirect RDMA, Kubernetes/custom gang schedulers, Hopper & Blackwell superclusters. Workplace: Onsite in San Francisco, CA. Compensation: $290,000 – $385,000 base salary plus flagship equity grants, comprehensive healthcare, and full relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of systems engineering experience in distributed ML training infrastructure, HPC supercomputing runtimes, or GPU collective communications.\n• Deep production expertise in PyTorch distributed internals, Megatron-Core/FSDP, C++17/20, CUDA streams/events, and NCCL / RDMA networking mechanics.\n• Proven track record diagnosing stragglers, deadlocks, memory fragmentation, and numerical divergence on multi-hundred- to multi-thousand-GPU training jobs.\n• Strong command of GPU microarchitecture, HBM/NVLink/NVSwitch bandwidth hierarchies, and PCIe/NIC topology alignment.\n• Demonstrated technical leadership driving cross-functional architecture across research, kernel, networking, and cluster operations teams.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience operating or writing training runtimes on 4,096+ H100/B200 GPU clusters.\n• Contributions to NCCL, PyTorch Distributed, Megatron-LM, DeepSpeed, TorchTitan, or custom RDMA transport libraries.\n• Experience with asynchronous/elastic RL training architectures or disaggregated pre-training storage fabrics.",
      "updated_at": "2026-11-15T17:37:18.217Z",
      "created_at": "2026-11-15T16:49:17.103Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-distributed-training-systems-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-distributed-training-systems-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-distributed-training-systems-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-distributed-training-systems-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "03d77aaf-de94-4549-a4fd-8d2c8264933e",
      "slug": "research-engineer-post-training-reinforcement",
      "title": "Research Engineer, Post-Training & Reinforcement Learning",
      "department": "AI Research",
      "department_code": "RES",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 265000,
        "max": 370000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$265k – $370k",
        "subtitle": "annual package"
      },
      "summary": "Research Engineer, Post-Training & Reinforcement Learning (AI Research) — San Francisco, CA · On-site · $265k – $370k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of the world's flagship frontier artificial intelligence research and deployment companies, headquartered in San Francisco, CA and serving hundreds of millions of global users and enterprise developers. Their Post-Training & Reinforcement Learning organization drives the reasoning, coding, mathematical problem-solving, and multimodal tool-use breakthroughs behind their flagship frontier models.\n\nThey have retained Talent247 to appoint a Research Engineer for Post-Training & Reinforcement Learning onsite in San Francisco. Embedded directly inside a tight-knit frontier post-training pod, the successful candidate will co-design large-scale online reinforcement learning algorithms, verifiable reasoning curricula, and asynchronous multi-thousand-GPU training infrastructure.\n\nKEY RESPONSIBILITIES\n• Design, implement, and scale online reinforcement learning runs—spanning PPO, GRPO, and novel off-policy reasoning algorithms—across multi-hundred-billion-parameter dense and Mixture-of-Experts (MoE) architectures.\n• Architect high-throughput distributed rollout generation and verifier execution pipelines pairing low-latency inference engines with FSDP/Megatron training sharding at >50% MFU.\n• Construct verifiable and rubric-guided reward environments across competitive programming, multi-file software engineering, formal mathematics, scientific synthesis, and multi-turn agentic tool execution.\n• Diagnose and eliminate subtle post-training pathologies including entropy collapse, length hacking, sycophancy, verifier exploitation, and catastrophic forgetting of pre-trained world knowledge.\n• Run disciplined small-to-large scaling-law ablations that forecast test-time compute efficiency and downstream frontier benchmark jumps before committing cluster-scale compute.\n• Partner directly with pre-training, inference kernel, and safety alignment teams through flagship model release cycles.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Triton/CUDA, custom distributed RL actor-learner frameworks, Megatron/FSDP, Kubernetes, multi-thousand H100/B200 GPU superclusters. Workplace: Onsite in San Francisco, CA. Compensation: $265,000 – $370,000 base salary plus top-decile frontier lab equity, comprehensive medical coverage, daily meals, and full relocation/visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of research engineering or ML systems experience (or PhD in Machine Learning/CS) with direct ownership of large-scale LLM post-training or reinforcement learning pipelines.\n• Deep production mastery of PyTorch internals, distributed parallelism (TP, PP, EP, CP, FSDP), and numerical stability debugging in BF16/FP8 regimes.\n• Hands-on track record implementing RLHF, RLAIF, PPO, GRPO, DPO, or verifiable-reward RL curricula on transformer architectures.\n• Strong systems programming ability to profile GPU kernel bottlenecks, NCCL collective stalls, and asynchronous rollout-to-learner network throughput.\n• High experimental velocity and scientific rigor when isolating subtle algorithmic gains in noisy RL training curves.\n\nPREFERRED / BONUS EXPERIENCE\n• Core contributions to shipped frontier reasoning models, competitive-coding/math RL breakthroughs, or major open-source RLHF frameworks.\n• Experience co-designing continuous-batching inference engines (vLLM/SGLang/TensorRT-LLM) for on-policy RL rollout generation.\n• Publications at NeurIPS, ICML, ICLR, or ACL, or strong competitive programming / Olympiad background.",
      "updated_at": "2026-11-15T17:37:09.970Z",
      "created_at": "2026-11-15T16:49:14.640Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-engineer-post-training-reinforcement",
        "markdown": "https://jobs.talent247.co/jobs/research-engineer-post-training-reinforcement.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-engineer-post-training-reinforcement/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-engineer-post-training-reinforcement/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "e614a7b7-9e00-43eb-bb5b-97105a60a693",
      "slug": "senior-research-scientist-autonomous-scientific-ai",
      "title": "Senior Research Scientist, Autonomous Scientific AI",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Cambridge, MA (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 220000,
        "max": 310000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$220k – $310k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Scientist, Autonomous Scientific AI (AI Research) — Cambridge, MA · On-site · $220k – $310k. THE CLIENT & MANDATE Our client is a heavily…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a heavily capitalized ($500M+ raised) scientific superintelligence pioneer headquartered in Cambridge, MA, founded out of a premier biotechnology ecosystem to reimagine the scientific method itself. Rather than training passive literature-summarization models, they couple frontier multimodal scientific foundation models directly to autonomous robotic wet laboratories spanning life sciences, chemistry, and sustainable materials discovery.\n\nThey have retained Talent247 to appoint a Senior Research Scientist in Autonomous Scientific AI onsite in Cambridge, MA. Collaborating directly with experimental biologists, synthetic chemists, and robotics engineers, the successful candidate will design closed-loop foundation models and Bayesian active-learning policies where every model hypothesis is physically synthesized, assayed, and fed back into training within days.\n\nKEY RESPONSIBILITIES\n• Architect multimodal scientific foundation models uniting protein/genomic sequences, small-molecule and crystal graphs, spectroscopic/microscopy readouts, and natural-language experimental protocols.\n• Formulate epistemic uncertainty estimation, multi-fidelity surrogate modeling, and batch Bayesian active-learning algorithms that maximize mutual information per robotic wet-lab plate run.\n• Pioneer reinforcement learning and causal-discovery post-training loops where real physical experimental outcomes—including negative results and unexpected reaction pathways—serve as verifiable ground-truth rewards.\n• Design cross-domain representation learning frameworks transferring physical and chemical priors across catalysis, carbon-capture materials, nucleic acid delivery, and protein engineering.\n• Partner side-by-side with wet-lab automation leads in Cambridge to diagnose assay noise, batch effects, and instrument drift directly inside probabilistic observation models.\n• Publish landmark scientific discoveries and machine learning advances while shaping multi-year compute and experimental data generation roadmaps.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, JAX, geometric & sequence transformers, BoTorch/Gaussian Processes/ensemble uncertainty, RDKit/Biopython, multi-node H100 GPU clusters integrated with automated robotic wet labs. Workplace: Onsite in Cambridge, MA (Kendall Square). Compensation: $220,000 – $310,000 base salary plus high-upside equity ownership, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computational Biology, Computational Chemistry, Chemical Engineering, Materials Science, or Physics with 3+ years of post-doctoral or frontier industrial AI research experience.\n• Strong publication or invention record in AI for science, molecular/biological foundation models, active learning, Bayesian optimization, or closed-loop autonomous experimentation.\n• Deep hands-on proficiency in PyTorch or JAX for training custom multimodal, graph, or sequence architectures on multi-GPU clusters.\n• Rigorous grounding in experimental design, calibrated uncertainty quantification under heavy distribution shift, and noisy physical measurement modeling.\n• Enthusiasm for working onsite alongside live automated laboratories where algorithms steer physical experiments every day.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience closing the loop between ML models and high-throughput wet-lab screening (high-content imaging, NGS/omics, DELs, electrochemistry, or automated synthesis).\n• Expertise in flow matching / diffusion for molecular or materials generation, or fine-tuning reasoning LLMs for scientific hypothesis generation.\n• Track record bridging ML research with wet-lab scientists in a fast-scaling biotech or frontier AI research organization.",
      "updated_at": "2026-11-15T17:36:04.251Z",
      "created_at": "2026-11-15T16:52:18.857Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-scientist-autonomous-scientific-ai",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-scientist-autonomous-scientific-ai.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-autonomous-scientific-ai/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-autonomous-scientific-ai/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "c8f7b8d3-d120-471f-abff-798adb03de76",
      "slug": "senior-ai-supercomputing-sre-contract",
      "title": "Senior AI Supercomputing SRE (Contract)",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Austin, TX (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Contract",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 180000,
        "max": 235000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$180k – $235k",
        "subtitle": "annual package"
      },
      "summary": "Senior AI Supercomputing SRE (Contract) (MLOps & Platform) — Austin, TX · On-site · $180k – $235k. THE CLIENT & MANDATE Our client is a leading custom AI…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a leading custom AI silicon and supercomputing company scaling multi-megawatt liquid-cooled AI inference and training datacenters, anchored by a core operations and infrastructure engineering center in Austin, TX. Operating tightly coupled bare-metal accelerator clusters requires deep HPC systems craftsmanship spanning firmware, high-radix 100G/400G Ethernet fabrics, and automated cluster self-healing.\n\nThey have partnered with Talent247 to appoint a Senior AI Supercomputing SRE on a 12-month contract (with strong conversion/extension potential) onsite in Austin, TX. Embedded directly with cluster bring-up and production cloud reliability teams, the post-holder will automate bare-metal provisioning, telemetry, and fault remediation across next-generation AI supercomputers.\n\nKEY RESPONSIBILITIES\n• Automate bare-metal OS provisioning (PXE/iPXE, Tinkerbell/MAAS, Ansible), firmware/BIOS/BMC lifecycle upgrades via Redfish, and Kubernetes/Slurm node onboarding across multi-rack AI supercomputing pods.\n• Build real-time hardware health daemons and automated cordoning/remediation workflows detecting optical transceiver degradation, ECC memory errors, PSU faults, and liquid-cooling thermal/flow anomalies.\n• Manage and troubleshoot lossless 100G/400G RoCEv2 / L3 CLOS datacenter fabrics (PFC, ECN, BGP EVPN, LLDP) connecting host CPU nodes, Ceph/NVMe storage arrays, and custom AI silicon systems.\n• Construct unified Prometheus, VictoriaMetrics, Grafana, and Loki observability stacks correlating facility CDU/PDU power-thermal metrics directly with training/inference job SLOs.\n• Execute structured burn-in validation, synthetic stress benchmarks, and acceptance sign-offs on newly delivered supercomputer racks to compress rack-to-revenue timelines from weeks to hours.\n• Author crisp runbooks, post-incident root-cause analyses, and self-service diagnostic CLI tools for firmware and runtime engineers.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Linux (Ubuntu/RHEL kernel tuning), Python, Go, Bash, Kubernetes (bare-metal), Slurm, Ansible/Terraform, Redfish/IPMI/OpenBMC, 100G/400G Arista/SONiC RoCEv2, Prometheus/Grafana, NVMe-oF/Ceph. Workplace: Onsite in Austin, TX (supporting live supercomputing clusters and lab pods). Compensation: $180,000 – $235,000 annualized contract equivalent ($90–$120/hr W-2 or Corp-to-Corp) with long-term extension or full-time conversion pathway.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 6+ years of HPC systems engineering, bare-metal AI infrastructure SRE, or hyperscale datacenter fleet automation experience.\n• Deep hands-on mastery of Linux internals (PCIe topology, NUMA pinning, hugepages, kernel modules, systemd, network stack tuning) and out-of-band BMC/Redfish automation.\n• Proven production experience administering bare-metal Kubernetes or Slurm clusters alongside Infrastructure-as-Code tooling (Ansible, Terraform) and Python/Go automation.\n• Practical knowledge of high-speed datacenter networking (100G/400G Ethernet, BGP, RoCEv2/RDMA, optical link debugging) and high-performance parallel/NVMe storage.\n• Calm, methodical troubleshooting skills across hardware, firmware, network, and container layers.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct hands-on experience operating direct-to-chip or rear-door liquid-cooled AI racks, high-density PDUs, and datacenter DCIM telemetry integrations.\n• Background supporting large GPU (H100/B200) or custom AI accelerator supercomputing clusters at an AI lab, neocloud, or national supercomputing center.\n• Familiarity with SONiC network switches, eBPF system diagnostics, or Ceph/WEKA/Lustre parallel filesystems.",
      "updated_at": "2026-11-15T17:36:03.464Z",
      "created_at": "2026-11-15T16:52:18.187Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ai-supercomputing-sre-contract",
        "markdown": "https://jobs.talent247.co/jobs/senior-ai-supercomputing-sre-contract.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ai-supercomputing-sre-contract/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ai-supercomputing-sre-contract/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "6c548f43-5375-451c-8e42-884adcecc373",
      "slug": "staff-ml-compiler-engineer-custom-ai-silicon",
      "title": "Staff ML Compiler Engineer, Custom AI Silicon",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Sunnyvale, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 245000,
        "max": 345000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$245k – $345k",
        "subtitle": "annual package"
      },
      "summary": "Staff ML Compiler Engineer, Custom AI Silicon (AI Infrastructure) — Sunnyvale, CA · Hybrid · $245k – $345k. THE CLIENT & MANDATE Our client is an iconic…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an iconic Silicon Valley AI hardware and supercomputing pioneer headquartered in Sunnyvale, CA, designing radical custom AI silicon and ultra-high-speed inference and training clusters. By eliminating traditional off-chip memory bottlenecks through massive on-chip SRAM and spatial dataflow fabric architectures, their systems deliver industry-leading token generation speeds for frontier reasoning and coding models.\n\nThey have retained Talent247 to appoint a Staff ML Compiler Engineer for Custom AI Silicon in Sunnyvale. Sitting at the heart of their compiler stack, the successful candidate will own MLIR/LLVM graph lowering, spatial dataflow placement, and SRAM tiling algorithms that unlock peak hardware utilization for 100B+ Mixture-of-Experts (MoE) and long-context transformers.\n\nKEY RESPONSIBILITIES\n• Architect multi-level MLIR and C++20 compiler passes that lower PyTorch/JAX computation graphs (Torch-MLIR / StableHLO) onto massively parallel 2D spatial dataflow processing elements.\n• Design polyhedral loop scheduling, tensor sharding, and on-chip SRAM memory-planning algorithms that overlap compute with fabric-level collective communications at near-zero idle overhead.\n• Pioneer compiler optimizations for sparse Mixture-of-Experts (MoE) dynamic routing, multi-token speculative decoding, Flash/linear attention variants, and sub-8-bit quantized arithmetic (FP8/INT4/MXFP).\n• Build cycle-accurate performance estimators and automated autotuning frameworks guiding operator fusion, layout selection, and spatial core placement across hundreds of thousands of cores.\n• Partner directly with silicon microarchitects to co-design next-generation instruction sets, collective routing primitives, and sparsity hardware units for upcoming tape-outs.\n• Provide technical leadership across compiler engineers and kernel authors to deliver day-zero support for newly released open-weight and frontier model architectures.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Modern C++20, MLIR, LLVM, Python, PyTorch 2.x (TorchDynamo/Inductor), StableHLO/ONNX, polyhedral/ILP solvers, cycle-accurate hardware simulators & live silicon clusters. Workplace: Hybrid in Sunnyvale, CA. Compensation: $245,000 – $345,000 base salary plus substantial equity grants, performance bonuses, and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of compiler engineering experience (or PhD in Compilers/Computer Architecture + 4+ years industry impact) working on ML compilers, HPC code generation, or GPU/ASIC toolchains.\n• Deep production mastery of MLIR and LLVM internals in modern C++ (dialect design, pattern rewriting, bufferization, affine/linalg lowering, register/memory allocation).\n• Strong command of deep learning operator semantics (attention variants, MoE routing, KV-cache management, collective communications) and hardware roof-line modeling.\n• Experience with spatial dataflow architectures, systolic arrays, scratchpad/SRAM-managed accelerators, or GPU warp/tensor-core kernel compilation (Triton, XLA, TVM, IREE).\n• Proven ability to diagnose numerical precision issues (BF16/FP8/FP4) and performance cliffs across full end-to-end 70B–600B+ parameter models.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience co-designing compiler stacks for custom AI ASICs, TPUs, NPUs, or reconfigurable dataflow processors.\n• Expertise in integer linear programming (ILP), constraint satisfaction, or RL/search-based graph placement and routing algorithms.\n• Core upstream contributions to MLIR, LLVM, OpenXLA, Triton, or PyTorch compiler subsystems.",
      "updated_at": "2026-11-15T17:35:43.969Z",
      "created_at": "2026-11-15T16:52:11.730Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-ml-compiler-engineer-custom-ai-silicon",
        "markdown": "https://jobs.talent247.co/jobs/staff-ml-compiler-engineer-custom-ai-silicon.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-ml-compiler-engineer-custom-ai-silicon/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-ml-compiler-engineer-custom-ai-silicon/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "2246eb7e-2854-408a-817d-eb8a96584daa",
      "slug": "senior-backend-engineer-agent-connectors-mcp",
      "title": "Senior Backend Engineer, Agent Connectors & MCP Platform",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Seattle, WA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 190000,
        "max": 270000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$190k – $270k",
        "subtitle": "annual package"
      },
      "summary": "Senior Backend Engineer, Agent Connectors & MCP Platform (Applied AI) — Seattle, WA · Hybrid · $190k – $270k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier enterprise work-AI and knowledge platform scaling rapidly across Palo Alto and Seattle. As enterprise adoption evolves from read-only search toward autonomous cross-application action execution, their Agent Connectors & Tooling Platform team builds the secure execution fabric enabling AI agents to read and write safely across Salesforce, Jira, GitHub, Slack, ServiceNow, Workday, and custom internal APIs.\n\nThey have partnered with Talent247 to hire a Senior Backend Engineer for their Agent Connectors & MCP Platform in Seattle, WA. The successful candidate will architect universal Model Context Protocol (MCP) gateways, per-user OAuth2/OIDC token delegation brokers, and transactional human-approval guardrails for enterprise agent fleets.\n\nKEY RESPONSIBILITIES\n• Architect a high-throughput, multi-tenant Model Context Protocol (MCP) server and connector gateway exposing hundreds of third-party SaaS and internal enterprise APIs as typed, LLM-ergonomic tool schemas.\n• Engineer fine-grained delegated identity brokers (OAuth 2.1, OIDC, SAML, scoped on-behalf-of token exchange) ensuring agents strictly inherit the invoking human user's live permissions on every read and write call.\n• Build declarative policy engines, dry-run diff previews, idempotency keys, and asynchronous human-in-the-loop approval gates before agents execute mutating external actions.\n• Design adaptive per-tenant rate-limiters, circuit breakers, webhook/CDC incremental sync pipelines, and schema-drift detectors resilient to upstream SaaS API changes.\n• Collaborate with applied AI engineers to optimize tool descriptions, error-recovery hints, and pagination semantics so models achieve 99%+ first-try tool-call validity.\n• Deliver enterprise-grade SDKs and self-serve connector builders allowing Fortune 500 platform teams to register proprietary internal microservices in minutes.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Go, TypeScript, Python, gRPC, Model Context Protocol (MCP), OAuth 2.1 / OIDC / KMS, PostgreSQL, Redis, Kafka/PubSub, Kubernetes, GCP/AWS. Workplace: Hybrid in Seattle, WA. Compensation: $190,000 – $270,000 base salary plus generous late-stage equity and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of backend and distributed systems engineering experience building API platforms, integration fabrics, identity/security systems, or cloud infrastructure in Go, TypeScript, Java, or Rust.\n• Deep practical mastery of OAuth 2.0/2.1, OIDC, webhook architectures, distributed rate limiting, idempotency, and multi-tenant secret management.\n• Experience designing clean, strongly typed API schemas (OpenAPI, JSON Schema, Protobuf, or MCP) consumed by external developers or AI agents.\n• Proven track record operating high-reliability services on Kubernetes with comprehensive structured logging, distributed tracing, and tenant isolation.\n• Security-first engineering mindset accustomed to enterprise threat modeling (confused-deputy prevention, SSRF mitigation, audit logging).\n\nPREFERRED / BONUS EXPERIENCE\n• Direct hands-on experience building or hosting production Model Context Protocol (MCP) servers, agent tool sandboxes, or iPaaS connector engines.\n• Familiarity with how LLMs parse tool schemas, handle structured error recovery, and remain vulnerable to indirect prompt injection via untrusted tool outputs.\n• Prior experience integrating complex enterprise SaaS APIs (Atlassian, Salesforce, Microsoft Graph, GitHub, Google Workspace, ServiceNow).",
      "updated_at": "2026-11-15T17:35:32.002Z",
      "created_at": "2026-11-15T16:52:08.215Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-backend-engineer-agent-connectors-mcp",
        "markdown": "https://jobs.talent247.co/jobs/senior-backend-engineer-agent-connectors-mcp.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-backend-engineer-agent-connectors-mcp/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-backend-engineer-agent-connectors-mcp/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "9f20fd59-bc8a-446f-b4cf-e7ad3b5daa23",
      "slug": "senior-machine-learning-engineer-enterprise-search",
      "title": "Senior Machine Learning Engineer, Enterprise Search & RAG",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Palo Alto, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 210000,
        "max": 295000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$210k – $295k",
        "subtitle": "annual package"
      },
      "summary": "Senior Machine Learning Engineer, Enterprise Search & RAG (Machine Learning) — Palo Alto, CA · Hybrid · $210k – $295k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar Silicon Valley enterprise AI search and work-agent leader headquartered in Palo Alto, CA. Deployed across hundreds of global enterprises, their platform indexes billions of documents, code repositories, tickets, and chat threads into a real-time, permission-aware enterprise knowledge graph powering grounded assistants and autonomous department agents.\n\nThey have retained Talent247 to appoint a Senior Machine Learning Engineer for Enterprise Search & RAG in Palo Alto. Working on their core retrieval and ranking engine, the successful candidate will advance personalized dense-lexical retrieval, multi-hop query planning, and cross-encoder reranking subject to strict real-time access-control enforcement.\n\nKEY RESPONSIBILITIES\n• Train and iterate on domain-adaptive bi-encoder embedding models, late-interaction (ColBERT-style) retrievers, and low-latency cross-encoder rerankers tailored to heterogeneous enterprise SaaS corpora.\n• Design agentic multi-hop query decomposition and iterative retrieval policies that resolve organizational acronyms, people-graph relationships, temporal recency, and code-to-ticket dependencies.\n• Co-design retrieval scoring and candidate-generation stages alongside search systems engineers so real-time user-level ACL filtering across 100+ SaaS sources never degrades recall or sub-200ms p99 latency.\n• Build privacy-preserving online learning-to-rank and implicit-feedback pipelines (clickstreams, citation dwell, reformulations, thumbs feedback) across strictly isolated single-tenant deployments.\n• Develop calibrated context-selection, dynamic snippet compression, and source-authority ranking algorithms that maximize downstream LLM answer faithfulness.\n• Establish rigorous offline counterfactual evaluation and interleaved online A/B testing frameworks guiding weekly ranking launches.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Go/C++, custom HNSW/inverted-index search engine, ONNX/TensorRT, Ray, Kubernetes, GCP/AWS. Workplace: Hybrid in Palo Alto, CA. Compensation: $210,000 – $295,000 base salary plus high-value equity grants, comprehensive healthcare, daily meals, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of machine learning engineering experience in large-scale search ranking, information retrieval, neural RAG, or recommendation systems.\n• Deep hands-on track record training contrastive dense embeddings, SPLADE/learned sparse representations, or cross-encoder rerankers in PyTorch.\n• Strong command of classical IR fundamentals (BM25, inverted indices, query understanding, entity linking, temporal decay, learning-to-rank).\n• Experience optimizing neural ranking inference latency via model distillation, INT8/FP8 quantization, ONNX/TensorRT serving, and multi-stage cascade pruning.\n• Proven ability to diagnose ranking failures on skewed, permission-partitioned corpora without violating customer zero-access data boundaries.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience combining graph/social signals (org charts, authorship, collaboration graphs) with neural semantic retrieval.\n• Prior tenure on a large-scale web search, code search, or enterprise workspace search ranking team.\n• Publications at SIGIR, WSDM, EMNLP, ACL, KDD, or NeurIPS in information retrieval or retrieval-augmented generation.",
      "updated_at": "2026-11-15T17:35:21.739Z",
      "created_at": "2026-11-15T16:52:03.859Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-enterprise-search",
        "markdown": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-enterprise-search.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-enterprise-search/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-enterprise-search/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "c68e6cc6-e81b-438e-8ba6-4ea3fb7d45fd",
      "slug": "senior-multimodal-document-intelligence-engineer",
      "title": "Senior Multimodal Document Intelligence Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 200000,
        "max": 285000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$200k – $285k",
        "subtitle": "annual package"
      },
      "summary": "Senior Multimodal Document Intelligence Engineer (Applied AI) — New York, NY · Hybrid · $200k – $285k. THE CLIENT & MANDATE Our client is a flagship…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a flagship enterprise AI company serving elite global law firms, private equity sponsors, and corporate legal teams from New York and San Francisco. Every downstream reasoning workflow in their platform depends on lossless ingestion of tens of millions of complex real-world documents—ranging from heavily redacted scanned PDFs and multi-page nested cap-table spreadsheets to handwritten margin annotations and cross-referenced deal exhibits.\n\nThey have engaged Talent247 to recruit a Senior Multimodal Document Intelligence Engineer in New York, NY. Sitting at the intersection of vision-language modeling and high-throughput ingestion infrastructure, the successful candidate will build layout-aware VLM parsing pipelines and hierarchical document graphs that preserve every footnote, clause boundary, and tabular cell with near-perfect fidelity.\n\nKEY RESPONSIBILITIES\n• Train, fine-tune, and deploy compact Vision-Language Models (VLMs) and layout-detection transformers specialized for complex multi-column agreements, dense financial tables, redlines/track-changes, and degraded scans.\n• Architect hierarchical structural chunking and cross-reference resolution graphs that link defined terms, schedules, signature pages, and embedded exhibits across multi-thousand-page deal rooms.\n• Engineer GPU-accelerated ingestion pipelines (vLLM / TensorRT-LLM + async Rust/Python workers) capable of parsing tens of thousands of uploaded pages within seconds of data-room ingestion.\n• Build multimodal bounding-box grounding pipelines so every generated legal or financial claim highlights the exact pixel coordinates and table cell in the source document viewer.\n• Construct adversarial document-parsing benchmarks covering rotated scans, multi-page spanning tables, watermarks, multilingual appendices, and complex mathematical formulas.\n• Optimize cost-quality routing between ultra-fast classical OCR/PDF-tree heuristics and heavy VLM visual-token passes on ambiguous pages.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust, PyTorch, Vision-Language Models (Qwen-VL/Pixtral-class custom fine-tunes), vLLM, PDFium/Poppler, Apache Arrow/Postgres, Kubernetes. Workplace: Hybrid in New York, NY. Compensation: $200,000 – $285,000 base salary plus significant equity ownership and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of ML engineering experience in multimodal vision-language models, document AI, OCR/layout understanding, or visual information extraction.\n• Proven experience fine-tuning and serving VLMs, multimodal transformers, or layout-aware vision encoders in PyTorch and vLLM/TensorRT.\n• Deep familiarity with PDF internals, coordinate bounding-box alignment, complex table-structure recognition (HTML/Markdown/JSON serialization), and reading-order reconstruction.\n• Strong systems engineering skills to scale batch and real-time GPU document pipelines with tight p95 latency and memory budgets.\n• Disciplined evaluation mindset building cell-level TEDS, edit-distance, and downstream retrieval-recall benchmarks.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience distilling frontier multimodal models into sub-7B vision-language parsers running at hundreds of pages per second per GPU.\n• Background processing financial filings (10-K/S-1s), M&A virtual data rooms, clinical trial PDFs, or patent specifications.\n• Proficiency in Rust or C++ for high-speed PDF rasterization and structural pre-parsing.",
      "updated_at": "2026-11-15T17:35:10.022Z",
      "created_at": "2026-11-15T16:52:00.472Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-multimodal-document-intelligence-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-multimodal-document-intelligence-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-multimodal-document-intelligence-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-multimodal-document-intelligence-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "9b4c2e78-3e7f-4090-a9a1-6239763cb0d3",
      "slug": "senior-machine-learning-engineer-long-context",
      "title": "Senior Machine Learning Engineer, Long-Context Domain Post-Training",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 225000,
        "max": 315000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $315k",
        "subtitle": "annual package"
      },
      "summary": "Senior Machine Learning Engineer, Long-Context Domain Post-Training (Machine Learning) — San Francisco, CA · Hybrid · $225k – $315k. THE CLIENT & MANDATE…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is the category-defining domain AI platform serving the world's top AmLaw 100 firms, global investment banks, and Fortune 500 enterprises from headquarters in San Francisco. Unlike thin prompt wrappers, their core technical moat includes an internal Models & Post-Training research team that trains custom long-context reasoning models specifically optimized for multi-document legal, tax, and regulatory logic.\n\nThey have partnered with Talent247 to hire a Senior Machine Learning Engineer for Long-Context Domain Post-Training in San Francisco. Collaborating closely with domain experts (attorneys, tax specialists, M&A analysts) and frontier lab partners, the appointed engineer will advance rubric-guided reinforcement learning and 128k–1M+ token post-training recipes over semi-verifiable professional tasks.\n\nKEY RESPONSIBILITIES\n• Design and execute end-to-end supervised fine-tuning (SFT), preference optimization, and online reinforcement learning (GRPO/PPO/RLOO) runs across open-weight and partner frontier model families.\n• Pioneer rubric-based reward modeling and decomposed generative verifiers for semi-verifiable tasks (contract redlining, statutory conflict analysis, multi-exhibit cross-referencing) where simple unit tests do not exist.\n• Optimize long-context (128k–512k+ token) multi-document reasoning performance, eliminating lost-in-the-middle recall drop-offs, subtle citation fabrication, and premature summarization shortcuts.\n• Build synthetic trajectory generation, expert critique-and-revision loops, and hard-negative data mining pipelines seeded by in-house legal and financial specialists.\n• Partner with the agentic runtime team to co-train models directly inside live multi-step search, citation, and document-editing tool loops.\n• Maintain rigorous contamination-free domain benchmarks measuring win rates against unadapted frontier reasoning models.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Verl / OpenRLHF / Megatron-Core / FSDP2, vLLM/SGLang, FlashAttention-3 / Ring Attention, multi-node H100/B200 GPU clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $225,000 – $315,000 base salary plus high-value equity grants, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of hands-on machine learning engineering or research engineering experience focused on LLM post-training, alignment, or fine-tuning at scale.\n• Proven track record running distributed SFT, DPO, PPO, or GRPO training jobs on 30B+ or MoE transformer models using PyTorch, FSDP, DeepSpeed, or Megatron.\n• Practical expertise in reward model design, rubric/judge calibration, reward-hacking mitigation, and synthetic reasoning trace curation.\n• Deep familiarity with long-context training mechanics (RoPE scaling, context parallelism / Ring Attention, sequence packing, KV-cache memory profiling).\n• Strong experimental rigor—designing clean small-scale ablations that reliably predict large-scale post-training gains.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience training multi-turn tool-using or RAG-integrated reasoning agents via online reinforcement learning.\n• Publications or technical blog posts on RLHF/RLAIF, Constitutional AI, long-context needle-in-a-haystack reasoning, or domain adaptation.\n• Exposure to high-precision professional domains (law, finance, medicine, tax, or scientific literature).",
      "updated_at": "2026-11-15T17:35:00.676Z",
      "created_at": "2026-11-15T16:51:57.329Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-long-context",
        "markdown": "https://jobs.talent247.co/jobs/senior-machine-learning-engineer-long-context.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-long-context/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-machine-learning-engineer-long-context/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "739c9ca4-18c1-45e2-a47b-fd1db94f64d7",
      "slug": "staff-agentic-workflows-engineer-enterprise",
      "title": "Staff Agentic Workflows Engineer, Enterprise Reasoning",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 235000,
        "max": 325000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$235k – $325k",
        "subtitle": "annual package"
      },
      "summary": "Staff Agentic Workflows Engineer, Enterprise Reasoning (Applied AI) — New York, NY · Hybrid · $235k – $325k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar vertical AI leader transforming global law firms, corporate legal departments, private equity funds, and tax practices. From their New York and San Francisco engineering hubs, they build domain-specialized reasoning systems capable of executing multi-hour M&A due diligence, cross-jurisdictional regulatory analysis, and litigation drafting across thousands of interdependent source documents.\n\nThey have retained Talent247 to appoint a Staff Agentic Workflows Engineer for Enterprise Reasoning in New York, NY. Serving as an architectural lead on their core agentic platform, the successful candidate will design durable, multi-agent planning runtimes, verifiable citation engines, and adaptive context-compaction architectures operating at massive enterprise scale.\n\nKEY RESPONSIBILITIES\n• Architect durable, stateful multi-agent execution engines (built on Temporal, TypeScript, and Python) capable of decomposing complex legal and financial mandates into hundreds of parallel, fault-tolerant sub-tasks.\n• Design dynamic context-management, hierarchical scratchpad memory, and cross-document synthesis patterns that prevent reasoning degradation across 1,000+ document data rooms.\n• Engineer zero-hallucination span-level citation verification, contradiction detection, and structured tabular matrix generation pipelines audited by partner-level domain experts.\n• Build multi-model orchestration layers that dynamically route planning, deep verification, high-speed extraction, and drafting steps across custom post-trained models and frontier reasoning APIs.\n• Establish trace-level OpenTelemetry observability, replayable agent debugging sandboxes, and automated multi-turn evaluation gates catching latency or reasoning regressions prior to release.\n• Mentor senior backend and applied AI engineers while setting technical standards across tool-interface contracts, tenant isolation, and execution reliability.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: TypeScript, Python, Go, Temporal, PostgreSQL, hybrid vector/lexical search, multi-provider frontier & custom post-trained LLMs, Kubernetes, Azure/AWS. Workplace: Hybrid in New York, NY. Compensation: $235,000 – $325,000 base salary plus substantial late-stage unicorn equity and comprehensive benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of software and applied AI systems engineering experience, including Staff-level architectural ownership of high-scale distributed backend or agentic platforms.\n• Deep production expertise building multi-step LLM agent orchestrators, durable workflow state machines (Temporal or equivalent), tool-use protocols, and structured output validation.\n• Mastery of context engineering, hybrid retrieval orchestration, parallel map-reduce LLM execution patterns, and latency/cost optimization at high concurrency.\n• Production fluency in TypeScript and Python (plus Go or Rust advantageous) with strong API design and multi-tenant security discipline (SOC2, tenant KMS encryption, zero data retention).\n• Proven ability to partner closely with ML research teams to align runtime scaffolding with custom post-trained model capabilities.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience building mission-critical AI workflows in legal tech, quantitative finance, M&A data rooms, or enterprise search.\n• Hands-on experience designing evaluation frameworks for long-horizon, semi-verifiable multi-document agent trajectories.\n• Contributions to open-source agent runtimes, MCP server ecosystems, or durable execution tooling.",
      "updated_at": "2026-11-15T17:34:50.374Z",
      "created_at": "2026-11-15T16:51:54.788Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-agentic-workflows-engineer-enterprise",
        "markdown": "https://jobs.talent247.co/jobs/staff-agentic-workflows-engineer-enterprise.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-agentic-workflows-engineer-enterprise/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-agentic-workflows-engineer-enterprise/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3a06548d-1976-4b70-ac23-f80510e6f6ab",
      "slug": "research-engineer-robot-reinforcement-learning-sim",
      "title": "Research Engineer, Robot Reinforcement Learning & Sim-to-Real",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 235000,
        "max": 330000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$235k – $330k",
        "subtitle": "annual package"
      },
      "summary": "Research Engineer, Robot Reinforcement Learning & Sim-to-Real (Robotics & Embodied AI) — San Francisco, CA · On-site · $235k – $330k. THE CLIENT &…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a frontier physical-AI research laboratory in San Francisco developing generalist robot foundation models across diverse real-world manipulation and mobile tasks. While human teleoperation provides broad behavioral priors, achieving industrial-grade speed, contact-rich precision, and autonomous recovery from mistakes requires scalable reinforcement learning across GPU-accelerated simulation and autonomous real-world practice.\n\nThey have engaged Talent247 to recruit a Research Engineer specializing in Robot Reinforcement Learning & Sim-to-Real onsite in San Francisco. The successful candidate will build massively parallel GPU physics gyms and real-world offline-to-online RL pipelines that push pretrained VLA policies far beyond demonstration performance.\n\nKEY RESPONSIBILITIES\n• Architect real-world and hybrid reinforcement learning algorithms (offline-to-online RL, Q-chunking, advantage-weighted flow matching, residual RL) fine-tuning large VLA policies directly on robot fleets.\n• Build high-throughput GPU-accelerated physics simulation environments in MuJoCo MJX and NVIDIA Isaac Lab featuring automated real-to-sim asset reconstruction and contact-parameter system identification.\n• Design autonomous reset mechanisms, VLM-based dense success/failure reward detectors, and safe exploration controllers enabling unattended multi-hour real-robot RL training runs.\n• Engineer distributed actor-learner infrastructure streaming real-time multimodal rollouts from dozens of physical robot cells directly into centralized GPU training clusters.\n• Investigate domain randomization, tactile/force-conditioned policies, and privileged-critic asymmetric architectures for contact-rich assembly and deformable-object manipulation.\n• Collaborate closely with foundation model scientists to fold RL-refined high-skill trajectories back into generalist pre-training mixtures.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, MuJoCo / MJX, Isaac Lab / Warp, C++, distributed RL actor-learner frameworks, multi-arm physical robot fleet & H100 GPU clusters. Workplace: Onsite in San Francisco, CA. Compensation: $235,000 – $330,000 base salary plus high-upside equity and comprehensive benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• MS or PhD in Robotics, Machine Learning, or Computer Science (or 4+ years of research engineering experience) with a strong focus on robot RL and control.\n• Demonstrated success training RL policies that transfer to real physical robots (locomotion, dexterous manipulation, or contact-rich assembly) via sim-to-real or real-world RL.\n• Deep proficiency in JAX or PyTorch alongside GPU-parallel physics simulators such as MuJoCo MJX, Isaac Lab/Gym, or Brax.\n• Strong systems engineering ability to build low-overhead distributed rollout collection, replay buffers, and VLM reward-model pipelines.\n• Solid foundations in off-policy RL, actor-critic stability, value overestimation mitigation, and continuous action distributions.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience fine-tuning diffusion or flow-matching visuomotor policies with reinforcement learning (e.g., DPPO, IQL/Cal-QL, advantage-weighted regression, RLPD).\n• Track record of publications at CoRL, RSS, ICRA, IROS, NeurIPS, or ICML.\n• Hands-on familiarity with automated 3D scene/URDF reconstruction (3DGS/mesh-to-sim) for scalable task generation.",
      "updated_at": "2026-11-15T17:34:39.837Z",
      "created_at": "2026-11-15T16:51:51.753Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-engineer-robot-reinforcement-learning-sim",
        "markdown": "https://jobs.talent247.co/jobs/research-engineer-robot-reinforcement-learning-sim.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-engineer-robot-reinforcement-learning-sim/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-engineer-robot-reinforcement-learning-sim/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "df7a2c78-f6a1-4b31-9e84-5046d41d6b74",
      "slug": "senior-real-time-robotics-on-device-inference",
      "title": "Senior Real-Time Robotics & On-Device Inference Engineer",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 225000,
        "max": 320000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$225k – $320k",
        "subtitle": "annual package"
      },
      "summary": "Senior Real-Time Robotics & On-Device Inference Engineer (Robotics & Embodied AI) — San Francisco, CA · On-site · $225k – $320k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco generalist-robotics AI lab training multi-billion-parameter Vision-Language-Action (VLA) foundation models to perform complex dexterous manipulation in unstructured physical environments. Turning massive multimodal transformers into smooth, reactive 50Hz–100Hz bimanual robot motion requires tightly co-designed C++ real-time control loops and ultra-low-latency GPU inference.\n\nThey have partnered with Talent247 to hire a Senior Real-Time Robotics & On-Device Inference Engineer onsite in San Francisco. Collaborating directly with VLA model architects and hardware engineers, the appointed engineer will own the onboard execution stack bridging raw multi-camera streams, TensorRT/CUDA action-chunk inference, and low-level torque/impedance controllers.\n\nKEY RESPONSIBILITIES\n• Architect deterministic C++20 and CUDA/TensorRT inference engines running multi-billion-parameter VLA backbone + flow-matching action heads within <15ms end-to-end sensor-to-actuator latency.\n• Design asynchronous dual-rate execution architectures (decoupling heavy VLM KV-cache updates from high-frequency flow-matching action-chunk denoising and real-time trajectory interpolation).\n• Implement smooth temporal ensembling, receding-horizon spline blending, and compliant joint/Cartesian impedance controllers over CAN-FD, EtherCAT, and real-time Linux (`PREEMPT_RT`) buses.\n• Optimize custom CUDA/Triton kernels, CUDA Graphs, FP8/BF16 quantization, and zero-copy V4L2/CSI camera DMA pipelines on workstation and embedded edge GPUs.\n• Build high-precision hardware telemetry, safety envelopes (collision/torque watchdogs), and sub-millisecond synchronized episode recorders across dozens of daily robot stations.\n• Diagnose subtle sim-to-real and train-to-deploy discrepancies caused by camera exposure latency, clock skew, or numerical precision drift.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Modern C++20, CUDA, TensorRT, Python, JAX/PyTorch export (ONNX/TorchExport/StableHLO), Linux `PREEMPT_RT`, EtherCAT/CAN-FD, multi-arm & mobile manipulator hardware. Workplace: Onsite in San Francisco, CA. Compensation: $225,000 – $320,000 base salary plus meaningful equity ownership and full medical/401(k) benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of robotics systems, real-time control, or low-latency GPU inference engineering experience in modern C++ (C++17/20) and Python.\n• Proven experience optimizing deep neural network inference on NVIDIA GPUs using CUDA, TensorRT, CUDA Graphs, and mixed-precision quantization.\n• Strong command of real-time Linux systems (`PREEMPT_RT`), lock-free concurrency, hardware sensor synchronization, and robotic kinematics/dynamics.\n• Hands-on track record bringing up and debugging real physical robot arms, dexterous hands, or mobile manipulators.\n• Obsessive attention to p99.9 latency jitter, safety interlocks, and clean hardware abstraction interfaces.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience deploying diffusion policies, flow-matching action heads, or VLA transformers on bimanual manipulation platforms.\n• Familiarity with exporting JAX/XLA or PyTorch 2.x custom attention graphs cleanly to C++/TensorRT runtimes.\n• Experience with whole-body operational space control, force/tactile feedback integration, or custom high-bandwidth motor drivers.",
      "updated_at": "2026-11-15T17:34:30.168Z",
      "created_at": "2026-11-15T16:51:47.025Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-real-time-robotics-on-device-inference",
        "markdown": "https://jobs.talent247.co/jobs/senior-real-time-robotics-on-device-inference.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-real-time-robotics-on-device-inference/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-real-time-robotics-on-device-inference/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "119b09e1-7cb2-498c-9c43-3ffcfef4ec05",
      "slug": "research-scientist-vision-language-action",
      "title": "Research Scientist, Vision-Language-Action Foundation Models",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "San Francisco, CA",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 265000,
        "max": 375000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$265k – $375k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Vision-Language-Action Foundation Models (Robotics & Embodied AI) — San Francisco, CA · On-site · $265k – $375k. THE CLIENT & MANDATE…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of the world's most closely watched physical-intelligence research laboratories, headquartered in San Francisco and backed by top-tier Silicon Valley venture investors and AI pioneers. Their mission is to bring general-purpose foundation models into the physical world—training unified Vision-Language-Action (VLA) models capable of controlling diverse robotic morphologies across dexterous, deformable-object, and long-horizon real-world tasks.\n\nThey have retained Talent247 to appoint a Research Scientist specializing in Vision-Language-Action Foundation Models onsite in San Francisco. Working alongside co-founders and robot fleet teams, the successful candidate will advance pre-training and post-training architectures that bridge internet-scale semantic reasoning with continuous, high-frequency dexterous motor control.\n\nKEY RESPONSIBILITIES\n• Design and scale generalist Vision-Language-Action architectures uniting pretrained multimodal backbones with continuous flow-matching, diffusion, or autoregressive action-token heads.\n• Investigate cross-embodiment transfer representations enabling a single foundation policy to share physical intuition across single-arm, bimanual, mobile-manipulator, and dexterous-hand hardware.\n• Develop hierarchical planning-and-control paradigms where high-level semantic subgoal reasoning steers reactive 50Hz+ visuomotor execution and autonomous error recovery.\n• Pioneer post-training alignment and offline-to-online reinforcement learning recipes that turn imperfect teleoperation demonstrations into superhuman cycle speed and 99%+ task completion reliability.\n• Conduct rigorous scaling-law studies across heterogeneous data mixtures combining own-fleet teleoperation, open robotics corpora, egocentric human video, and synthetic simulation.\n• Validate every algorithmic advance directly on an expansive in-house fleet of physical robots operating continuously on site.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, flow-matching/diffusion action transformers, multi-node H100/B200 training clusters, large-scale in-house bimanual & mobile robot fleet. Workplace: Onsite in San Francisco, CA. Compensation: $265,000 – $375,000 base salary plus premier early-stage research lab equity and comprehensive benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Robotics, Machine Learning, Computer Vision, or equivalent frontier research track record in robot learning and multimodal foundation models.\n• First-author publications at RSS, CoRL, ICRA, NeurIPS, ICML, ICLR, or CVPR in visuomotor policy learning, VLA models, imitation learning, or generative control.\n• Expert proficiency in JAX or PyTorch across multi-node distributed training, custom attention masking, and continuous generative modeling (flow matching / diffusion).\n• Proven hands-on experience deploying learned neural policies onto real physical robot hardware—not purely simulated benchmarks.\n• Deep intuition for failure-mode debugging across camera placement, action chunking, proprioceptive conditioning, and data distribution shift.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior core contributions to landmark generalist robot policies, large-scale cross-embodiment datasets, or dexterous manipulation breakthroughs.\n• Experience combining VLM pre-training or egocentric video world models with downstream robotic control fine-tuning.\n• Strong systems mindset capable of co-designing model architectures around real-time 20ms–50ms onboard control-loop constraints.",
      "updated_at": "2026-11-15T17:34:20.564Z",
      "created_at": "2026-11-15T16:51:43.945Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-vision-language-action",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-vision-language-action.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-vision-language-action/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-vision-language-action/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "58ba8cfb-40fc-45c5-83b1-ac6d5d070b75",
      "slug": "senior-ai-data-platform-engineer-multimodal",
      "title": "Senior AI Data Platform Engineer, Multimodal Curation",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Seattle, WA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 190000,
        "max": 270000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$190k – $270k",
        "subtitle": "annual package"
      },
      "summary": "Senior AI Data Platform Engineer, Multimodal Curation (MLOps & Platform) — Seattle, WA · Hybrid · $190k – $270k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a flagship AI data infrastructure and post-training company powering the multimodal, video, and physical-AI datasets behind leading frontier model labs and robotics pioneers. From their Seattle engineering center, their Multimodal Data Platform group processes hundreds of petabytes of raw video, 3D LiDAR point clouds, multi-view audio, and bimanual robot teleoperation trajectories.\n\nThey have engaged Talent247 to appoint a Senior AI Data Platform Engineer for Multimodal Curation in Seattle, WA. The successful candidate will build distributed GPU-accelerated curation, semantic deduplication, and synthetic auto-labeling engines that transform noisy raw sensor captures into frontier-grade pre-training and post-training datasets.\n\nKEY RESPONSIBILITIES\n• Architect horizontal Ray Data and Spark pipelines running distributed GPU inference (VLM captioning, optical flow filtering, 3D bounding-box tracking, audio diarization) across petabyte multimodal lakes.\n• Build high-throughput columnar storage and random-access indexing layers on Lance, Apache Iceberg, and Parquet enabling sub-second spatio-temporal slice queries across billions of multimodal frames.\n• Implement semantic deduplication, aesthetic/motion quality scoring, and MinHash/embedding clustering algorithms that maximize downstream foundation model sample efficiency.\n• Engineer synchronized multi-sensor ingestion pipelines validating sub-millisecond timestamp alignment and camera-LiDAR-proprioception calibration across robotics and autonomous fleet uploads.\n• Design human-in-the-loop quality-routing engines that direct high-entropy model disagreements to expert annotators while auto-accepting high-confidence teacher labels.\n• Optimize spot-GPU scheduling, NVDEC hardware video decoding, and object-store multipart streaming to cut per-million-clip processing costs by 50%+.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust/C++, Ray Data, Apache Spark, Lance, Iceberg/Parquet/Arrow, PyTorch, vLLM, Kubernetes, S3/GCS. Workplace: Hybrid in Seattle, WA. Compensation: $190,000 – $270,000 base salary plus equity ownership and full US health & retirement benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of distributed data platform or ML infrastructure engineering experience handling multi-terabyte to multi-petabyte unstructured or multimodal datasets.\n• Production mastery of Ray Data, Apache Spark, or Dask in Python, paired with strong systems fluency in Rust, Go, or C++.\n• Hands-on experience with multimodal columnar table formats (Lance, Iceberg, Delta, Parquet) and GPU-accelerated batch inference orchestration.\n• Solid grasp of vector indexing, approximate nearest-neighbor clustering, and object-storage throughput optimization.\n• Track record owning end-to-end reliability, cost telemetry, and schema evolution for mission-critical data pipelines.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience processing robotics trajectory formats (RLDS, LeRobot, MCAP/ROS2 bags), 3D point clouds, or long-form multi-stream video.\n• Familiarity with hardware-accelerated video/image decoding (NVDEC, DALI) and VLM-based synthetic recaptioning pipelines.\n• Open-source contributions to Ray, Lance, Arrow, or multimodal dataset toolkits.",
      "updated_at": "2026-11-15T17:34:11.556Z",
      "created_at": "2026-11-15T16:51:40.195Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ai-data-platform-engineer-multimodal",
        "markdown": "https://jobs.talent247.co/jobs/senior-ai-data-platform-engineer-multimodal.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ai-data-platform-engineer-multimodal/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ai-data-platform-engineer-multimodal/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "98594e48-3f02-44db-8bcd-dccc6dbb06f7",
      "slug": "senior-applied-ai-engineer-mission-critical-agents",
      "title": "Senior Applied AI Engineer, Mission-Critical Agents",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "New York, NY (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 200000,
        "max": 280000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$200k – $280k",
        "subtitle": "annual package"
      },
      "summary": "Senior Applied AI Engineer, Mission-Critical Agents (Applied AI) — New York, NY · Hybrid · $200k – $280k. THE CLIENT & MANDATE Our client is a leading US…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a leading US AI platform and data infrastructure company headquartered in San Francisco with a rapidly growing Applied AI hub in New York City. Their Mission-Critical Agents division builds auditable, multi-agent decision-support and operational planning systems deployed inside high-consequence federal, defense, financial, and global logistics environments where hallucinations or unauthorized actions are unacceptable.\n\nThey have retained Talent247 to hire a Senior Applied AI Engineer for Mission-Critical Agents in New York, NY. Working directly with domain operators and core platform teams, the appointed engineer will build retrieval-grounded agent runtimes, deterministic verification guardrails, and portable self-hosted inference stacks capable of running across commercial clouds and air-gapped enclaves.\n\nKEY RESPONSIBILITIES\n• Architect stateful, multi-step agentic workflows combining hybrid multimodal retrieval, structured tool execution, human-in-the-loop approval gates, and immutable cryptographic audit logs.\n• Engineer zero-hallucination citation verification and schema-constrained decoding pipelines over heterogeneous intelligence reports, satellite/imagery metadata, financial filings, and operational databases.\n• Build portable, cloud-agnostic model-serving and orchestration stacks (Kubernetes, vLLM, quantized open-weight models) deployable cleanly into VPC-isolated and air-gapped environments.\n• Develop offline simulation replay harnesses and domain-specific regression suites that test agent robustness under degraded, contradictory, or adversarial input streams.\n• Fine-tune task-specialized tool-calling and reranking models to achieve frontier-grade reasoning accuracy within strict latency and on-prem GPU footprints.\n• Partner with customer mission leads in NYC and Washington, D.C. to translate high-stakes operational workflows into resilient production AI software.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, TypeScript, Go, PyTorch, vLLM/TensorRT-LLM, PostgreSQL/pgvector, OpenSearch, Temporal, Kubernetes, AWS/GovCloud/on-prem GPU clusters. Workplace: Hybrid in New York, NY. Compensation: $200,000 – $280,000 base salary plus equity grants, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of software and applied ML engineering experience building production backend systems and LLM/agent applications at scale.\n• Deep hands-on expertise in agentic orchestration (tool-use state machines, durable execution, context engineering), hybrid RAG, and structured output verification.\n• Practical experience deploying and optimizing open-weight LLMs/VLMs via vLLM, SGLang, or TensorRT-LLM inside Kubernetes environments.\n• Strong full-lifecycle engineering rigor in Python and TypeScript or Go, including observability, RBAC security enforcement, and automated evaluation gating.\n• US work authorization (US citizenship or permanent residency advantageous for select regulated customer engagements, though commercial tracks are also open).\n\nPREFERRED / BONUS EXPERIENCE\n• Experience deploying software into air-gapped, FedRAMP/IL4+, or highly regulated financial services environments.\n• Hands-on experience fine-tuning LoRA adapters or domain rerankers (SFT/DPO) on specialized technical or geospatial corpora.\n• Familiarity with Temporal workflow orchestration, Model Context Protocol (MCP) servers, and OpenTelemetry agent tracing.",
      "updated_at": "2026-11-15T17:34:00.502Z",
      "created_at": "2026-11-15T16:51:36.863Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-mission-critical-agents",
        "markdown": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-mission-critical-agents.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-mission-critical-agents/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-mission-critical-agents/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3c1be403-e298-437b-a513-31f18bbaae94",
      "slug": "machine-learning-engineer-frontier-safety-red-team",
      "title": "Machine Learning Engineer, Frontier Safety & Red-Team Evals",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 190000,
        "max": 270000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$190k – $270k",
        "subtitle": "annual package"
      },
      "summary": "Machine Learning Engineer, Frontier Safety & Red-Team Evals (Machine Learning) — San Francisco, CA · Hybrid · $190k – $270k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier San Francisco AI data, post-training, and evaluation leader operating independent frontier leaderboards and pre-deployment safety testing partnerships with leading commercial labs and national AI safety institutes. Their Frontier Safety & Evaluations group designs ungameable private benchmarks and automated red-teaming systems that measure catastrophic dual-use risks, autonomous cyber capabilities, and deceptive agent behaviors before public model release.\n\nThey have partnered with Talent247 to recruit a Machine Learning Engineer for Frontier Safety & Red-Team Evals in San Francisco. Combining psychometric rigor, adversarial ML, and sandboxed agent harnesses, the successful candidate will build the gold-standard safety measurement stack relied upon across the global AI ecosystem.\n\nKEY RESPONSIBILITIES\n• Architect private, contamination-resistant multimodal and agentic safety benchmarks evaluating cyber-offense uplift, biosecurity/chemical dual-use risk, persuasion, autonomous replication, and scheming.\n• Build multi-turn automated red-teaming swarms that adaptively discover multi-step jailbreaks, tool-use privilege escalations, and indirect prompt injections across frontier APIs and open-weight checkpoints.\n• Apply Item Response Theory (IRT), Bradley-Terry-Luce ranking models, and bootstrap confidence estimation to turn noisy human and automated red-team trials into statistically calibrated risk scores.\n• Engineer air-gapped, network-jailed Capture-the-Flag (CTF) and vulnerable-application sandboxes to safely measure autonomous vulnerability discovery and exploit generation.\n• Train specialized safety judge models and rubric verifiers calibrated against PhD domain-expert human adjudicators.\n• Co-author influential public technical reports and private pre-launch safety audits delivered to frontier model developers and regulatory partners.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, Inspect-style agent eval frameworks, Docker/microVM sandboxes, vLLM, statistical modeling (PyMC/Stan/SciPy), Kubernetes. Workplace: Hybrid in San Francisco, CA. Compensation: $190,000 – $270,000 base salary plus equity participation and comprehensive US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 3+ years of machine learning engineering, AI safety research engineering, or adversarial security ML experience.\n• Proven track record designing LLM/agent evaluation harnesses, adversarial red-teaming pipelines, or fine-tuning classifier/judge models in PyTorch.\n• Solid statistical grounding in benchmark design, inter-annotator agreement, variance reduction, contamination auditing, and ranking calibration.\n• Strong Python software engineering skills and familiarity with containerized sandbox isolation for untrusted model-generated code execution.\n• High integrity and security mindset suitable for handling sensitive pre-release frontier model weights and dual-use evaluation datasets.\n\nPREFERRED / BONUS EXPERIENCE\n• Background in offensive cybersecurity (CTF competitions, vulnerability research), computational biology/chemistry, or AI alignment evaluations.\n• Publications or open-source contributions in AI safety benchmarks, jailbreak robustness, or evaluation methodology.\n• Experience evaluating multimodal (voice/vision) or long-horizon computer-use agents.",
      "updated_at": "2026-11-15T17:33:51.447Z",
      "created_at": "2026-11-15T16:51:32.634Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/machine-learning-engineer-frontier-safety-red-team",
        "markdown": "https://jobs.talent247.co/jobs/machine-learning-engineer-frontier-safety-red-team.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-frontier-safety-red-team/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-frontier-safety-red-team/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "612023b1-7930-4069-842d-4028dd8f19b9",
      "slug": "senior-research-engineer-rl-reasoning-environments",
      "title": "Senior Research Engineer, RL Reasoning Environments",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "San Francisco, CA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 220000,
        "max": 310000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$220k – $310k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Engineer, RL Reasoning Environments (Machine Learning) — San Francisco, CA · Hybrid · $220k – $310k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar AI infrastructure and post-training powerhouse headquartered in San Francisco, supplying frontier AI laboratories with the verifiable reinforcement learning environments, human-expert reasoning traces, and evaluation harnesses behind today's top reasoning models. As post-training shifts from static preference ranking toward multi-turn RL with verifiable rewards (RLVR), their RL Environments group builds the interactive sandboxes where frontier models learn to code, prove theorems, and operate real software stacks.\n\nThey have retained Talent247 to hire a Senior Research Engineer for RL Reasoning Environments in San Francisco. The appointed engineer will architect massively parallel containerized execution gyms, anti-reward-hacking verifiers, and difficulty-adaptive curricula supporting millions of concurrent RL rollouts.\n\nKEY RESPONSIBILITIES\n• Architect deterministic, sub-second containerized RL environments spanning multi-repo software engineering, systems debugging, formal verification, quantitative finance, and browser/OS computer use.\n• Design tamper-resistant execution graders, mutation-tested unit suites, and hybrid formal/rubric reward verifiers impervious to reward hacking, test-file snooping, and environment side-channel leaks.\n• Run online reinforcement learning ablations (GRPO, PPO, RLOO) on open-weight and frontier models to validate that environment curricula produce genuine out-of-distribution capability transfer.\n• Build automated task-synthesis and repository-mining pipelines that transform real-world commit histories, bug trackers, and scientific workflows into contamination-free RL tasks calibrated at 5%–40% frontier pass@k.\n• Engineer distributed rollout schedulers on Kubernetes/Ray capable of orchestrating hundreds of thousands of isolated microVMs/containers with deterministic network mocking and state snapshotting.\n• Partner directly with research leads at partner frontier labs to diagnose trajectory failure modes and shape custom environment specifications.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust/Go, PyTorch, vLLM/SGLang, Verl/OpenRLHF, Docker/Firecracker/Kubernetes, Ray, multi-node H100 GPU clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $220,000 – $310,000 base salary plus high-growth equity, comprehensive healthcare, and 401(k).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of research engineering, ML infrastructure, or RL systems experience (or PhD in ML/CS with strong systems output).\n• Hands-on experience building reinforcement learning environments, SWE/coding agent benchmarks, or post-training RLVR/GRPO/PPO pipelines.\n• Strong systems programming skills in Python plus Go, Rust, or C++, with deep proficiency in Linux namespaces, container isolation, Firecracker/gVisor sandboxes, and distributed orchestration.\n• Practical understanding of reward hacking, pass@k scaling dynamics, data contamination detection, and trajectory-level credit assignment.\n• Ability to move rapidly from research hypothesis to hardened infrastructure executing millions of rollouts per week.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct authorship or core contributions to SWE-bench-style coding harnesses, terminal/OS agent gyms, or formal proof environments.\n• Experience optimizing asynchronous RL rollout-to-trainer pipelines using vLLM/SGLang and FSDP/Megatron.\n• Strong competitive programming, compilers, or security/sandboxing background.",
      "updated_at": "2026-11-15T17:33:41.383Z",
      "created_at": "2026-11-15T16:51:29.959Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7d9ad1e2-6fe1-4b81-9f1e-835d3b7c214f",
      "slug": "principal-forward-deployed-ai-architect",
      "title": "Principal Forward-Deployed AI Architect",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Remote (US)",
      "workplace_type": "Remote",
      "employment_type": "Full-time",
      "is_remote": true,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 205000,
        "max": 285000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$205k – $285k",
        "subtitle": "annual package"
      },
      "summary": "Principal Forward-Deployed AI Architect (Applied AI) — Remote (US) · Remote · $205k – $285k. THE CLIENT & MANDATE Our client is a premier enterprise data…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier enterprise data and artificial intelligence platform organization trusted by global leaders across financial services, healthcare, manufacturing, and retail. Their Forward-Deployed AI Architecture practice embeds Staff and Principal-caliber builders directly alongside C-suite technical leadership at Fortune 500 enterprises to turn proprietary petabyte data estates into production compound AI systems.\n\nThey have engaged Talent247 to appoint a Principal Forward-Deployed AI Architect (Remote, US). Rather than producing slideware, the successful candidate will architect and ship end-to-end multi-agent retrieval systems, domain-specific post-training pipelines, and automated LLM evaluation harnesses that survive strict enterprise latency, accuracy, and governance gates.\n\nKEY RESPONSIBILITIES\n• Lead the end-to-end technical architecture of flagship enterprise compound AI systems—combining hybrid vector/SQL retrieval, tool-calling agent graphs, and custom domain-adapted models.\n• Design domain-specific evaluation harnesses, golden benchmark datasets, and calibrated LLM judge pipelines that raise customer task accuracy from fragile prototypes (60%) to production reliability (95%+).\n• Implement continued pre-training, supervised fine-tuning (SFT), and preference optimization (DPO/RLVR) recipes over proprietary customer corpora where off-the-shelf frontier models fall short.\n• Architect strict row- and column-level permission inheritance, PII redaction guardrails, and audit-traceable citation lineage across enterprise RAG and agentic workflows.\n• Optimize token economics and end-to-end p95 latency via semantic routing, small-model distillation, prompt prefix structuring, and batch vs. streaming workload partitioning.\n• Feed recurring architectural bottlenecks from Fortune 500 deployments directly back to core engineering and research product roadmaps.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, SQL, Spark/Ray, vLLM, hybrid vector search, OpenTelemetry tracing, MCP/tool-use frameworks, AWS/Azure/GCP. Workplace: Remote across the United States (up to 20–25% strategic enterprise travel). Compensation: $205,000 – $285,000 base salary plus generous equity, performance incentives, and full remote US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 8+ years of hands-on experience across machine learning engineering, distributed data architecture, and production AI systems deployment, including 3+ years leading enterprise LLM/agent rollouts.\n• Deep practical mastery of compound AI design: hybrid retrieval (dense + lexical + structured SQL), query decomposition, reranking, evaluation-driven iteration, and fine-tuning (LoRA/full-parameter SFT/DPO).\n• Strong production Python and SQL engineering capability—comfortable writing reference production code alongside customer principal engineers.\n• Deep command of cloud security, VPC isolation, RBAC/ABAC data governance, and compliance requirements in regulated industries (finance, healthcare, public sector).\n• Executive presence capable of steering CTO/VP-Engineering stakeholders away from anti-patterns toward measurable, high-ROI architectures.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior Principal/Staff experience in a forward-deployed engineering, applied AI lab, or frontier model solutions architecture team.\n• Hands-on experience distilling frontier teacher models into fast 7B–32B open-weight models for high-volume enterprise inference.\n• Published technical reference architectures, open-source agent/eval tooling, or conference talks on production AI reliability.",
      "updated_at": "2026-11-15T17:33:32.192Z",
      "created_at": "2026-11-15T16:51:26.292Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/principal-forward-deployed-ai-architect",
        "markdown": "https://jobs.talent247.co/jobs/principal-forward-deployed-ai-architect.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/principal-forward-deployed-ai-architect/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/principal-forward-deployed-ai-architect/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8a366857-94cd-4115-98be-a950f5198c19",
      "slug": "senior-c-engine-engineer-ai-native-query-execution",
      "title": "Senior C++ Engine Engineer, AI-Native Query Execution",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Seattle, WA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 210000,
        "max": 295000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$210k – $295k",
        "subtitle": "annual package"
      },
      "summary": "Senior C++ Engine Engineer, AI-Native Query Execution (AI Infrastructure) — Seattle, WA · Hybrid · $210k – $295k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an industry-defining data and AI infrastructure leader processing exabytes of analytical and unstructured enterprise data daily. Their core execution engine team in Seattle builds next-generation vectorized C++ query engines that fuse relational columnar analytics directly with batch LLM inference, multimodal extraction, and vector similarity search inside unified execution plans.\n\nThey have partnered with Talent247 to hire a Senior C++ Engine Engineer for AI-Native Query Execution in Seattle, WA. Sitting at the intersection of database internals and modern AI systems, the appointed engineer will design SIMD-accelerated operators and asynchronous GPU inference pipelines that execute millions of AI transformations directly inside SQL and DataFrame plans.\n\nKEY RESPONSIBILITIES\n• Design and implement vectorized C++20 query execution operators that fuse relational filtering, joins, and aggregations with asynchronous batch LLM, embedding, and multimodal extraction calls.\n• Architect cost-based query optimizer rules that reorder predicates, deduplicate prompt prefixes, and schedule expensive AI operators alongside cheap columnar filters to cut compute cost by orders of magnitude.\n• Build high-throughput Apache Arrow and Parquet zero-copy serialization, adaptive batching, and token-budget rate-limiting buffers bridging CPU execution threads with remote/local GPU inference pools.\n• Optimize SIMD (AVX-512 / ARM NEON) kernels for hybrid BM25 lexical + quantized vector (PQ/HNSW/IVF) top-k retrieval directly over object-store-backed tables.\n• Engineer deterministic fault-tolerance, partial-batch checkpointing, and speculative retry mechanisms so multi-hour petabyte AI-SQL jobs survive transient endpoint throttling or node preemption.\n• Profile CPU cache misses, memory allocator contention, and async coroutine scheduling across thousand-node analytical clusters.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Modern C++20, SIMD (AVX-512/NEON), Apache Arrow, Parquet, Lance, Rust/Scala bindings, async RPC/gRPC, Linux perf/eBPF, multi-cloud object storage. Workplace: Hybrid in Seattle, WA. Compensation: $210,000 – $295,000 base salary plus substantial equity package and full US benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems engineering experience in modern C++ (C++17/20) building database execution engines, compilers, storage engines, or HPC frameworks.\n• Deep understanding of vectorized columnar execution, Volcano vs. push-based pipelines, cache-conscious data structures, SIMD intrinsics, and custom memory management.\n• Production familiarity with Apache Arrow, Parquet, ORC, or similar columnar memory/file formats and query optimization principles.\n• Strong concurrency and asynchronous systems skills (lock-free queues, C++20 coroutines/futures, high-throughput RPC clients).\n• Demonstrated rigor in micro-benchmarking, hardware performance counter analysis, and deterministic fuzz/correctness testing.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience integrating vector search (FAISS, HNSW, DiskANN) or batch LLM inference scheduling directly into analytical query engines (Velox, DuckDB, DataFusion, ClickHouse).\n• Knowledge of LLM serving mechanics (continuous batching, prefix caching, structured JSON decoding) and how query ordering impacts cache hit rates.\n• Contributions to major open-source database or columnar data projects.",
      "updated_at": "2026-11-15T17:33:23.476Z",
      "created_at": "2026-11-15T16:51:22.839Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-c-engine-engineer-ai-native-query-execution",
        "markdown": "https://jobs.talent247.co/jobs/senior-c-engine-engineer-ai-native-query-execution.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-c-engine-engineer-ai-native-query-execution/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-c-engine-engineer-ai-native-query-execution/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "29f6663e-ea0b-410e-a2c6-d59cdfe7497a",
      "slug": "senior-mlops-platform-engineer-model-feature",
      "title": "Senior MLOps Platform Engineer, Model & Feature Serving",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Seattle, WA (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 195000,
        "max": 275000,
        "currency": "USD",
        "frequency": "yearly",
        "formatted": "$195k – $275k",
        "subtitle": "annual package"
      },
      "summary": "Senior MLOps Platform Engineer, Model & Feature Serving (MLOps & Platform) — Seattle, WA · Hybrid · $195k – $275k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally scaled enterprise data and AI platform company powering mission-critical analytics, vector retrieval, and compound AI systems for thousands of Fortune 2000 organizations. Their Seattle AI Platform group owns the control plane and ultra-low-latency serving infrastructure behind custom language models, embedding endpoints, and real-time online feature stores.\n\nThey have retained Talent247 to appoint a Senior MLOps Platform Engineer for Model & Feature Serving in Seattle, WA. The successful candidate will build high-reliability multi-tenant control planes, zero-downtime canary routers, and sub-10ms online feature/vector lookup services supporting millions of requests per second across global cloud regions.\n\nKEY RESPONSIBILITIES\n• Architect and scale multi-tenant model-serving and online feature-serving data planes delivering <10ms p99 lookup latencies and 99.99% availability across AWS, GCP, and Azure.\n• Build intelligent gateway routing layers supporting token-aware load balancing, prefix-cache affinity, shadow traffic mirroring, and automated canary rollbacks for LLM and classical ML endpoints.\n• Engineer unified governance, lineage, and RBAC enforcement pipelines connecting training checkpoints, quantized model artifacts, vector indices, and live inference telemetry.\n• Design fast autoscaling controllers and warm-pool GPU/CPU provisioning engines that react to bursty enterprise agent workloads within seconds.\n• Implement streaming inference-log capture and drift-monitoring sinks feeding offline evaluation and continuous fine-tuning loops without impacting serving tail latency.\n• Drive operational excellence through chaos testing, multi-region failover automation, and crisp SLO instrumentation.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Go, Rust, Scala/Java, Python, Envoy, gRPC, Kubernetes, RocksDB/serverless key-value stores, Prometheus/OpenTelemetry, AWS/GCP/Azure. Workplace: Hybrid in Seattle, WA. Compensation: $195,000 – $275,000 base salary plus high-value equity grants, annual bonus/refreshers, and comprehensive US health & 401(k) benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of distributed systems, MLOps platform, or cloud infrastructure engineering experience operating high-QPS multi-tenant services.\n• Production mastery of Go, Rust, C++, or Scala/Java alongside strong Kubernetes controller/operator and Envoy/gRPC networking expertise.\n• Proven track record building model-serving platforms, online feature stores, real-time vector indices, or API gateways under strict sub-10ms to sub-50ms p99 SLAs.\n• Deep familiarity with multi-cloud IAM/security isolation, zero-downtime deployment patterns, and distributed state consensus.\n• Strong operational ownership across observability, capacity modeling, and root-cause performance debugging.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience orchestrating LLM serving runtimes (vLLM, TensorRT-LLM, SGLang, Triton Inference Server) and KV-cache-aware request routing.\n• Background working on storage-backed low-latency serving engines (RocksDB, Redis, Aerospike, Apache Arrow Flight).\n• Open-source contributions to Kubernetes, Envoy, Ray Serve, MLflow, or cloud-native CNCF projects.",
      "updated_at": "2026-11-15T17:33:13.816Z",
      "created_at": "2026-11-15T16:51:19.362Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-mlops-platform-engineer-model-feature",
        "markdown": "https://jobs.talent247.co/jobs/senior-mlops-platform-engineer-model-feature.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-mlops-platform-engineer-model-feature/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-mlops-platform-engineer-model-feature/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "978e1cc8-03fa-4953-891c-f74b816b7b8b",
      "slug": "senior-ml-engineer-ultra-scale-molecular-inference",
      "title": "Senior ML Engineer, Ultra-Scale Molecular Inference & Active Learning",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Cambridge, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 110000,
        "max": 155000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£110k – £155k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Engineer, Ultra-Scale Molecular Inference & Active Learning (Machine Learning) — Cambridge, UK · Hybrid · £110k – £155k. THE CLIENT & MANDATE…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a world-leading AI-driven therapeutics research organization expanding their computational engineering presence in Cambridge, UK. Their platform pairs frontier all-atom biomolecular foundation models with ultra-high-throughput generative chemistry to explore chemical space orders of magnitude faster than traditional screening.\n\nThey have engaged Talent247 to hire a Senior ML Engineer for Ultra-Scale Molecular Inference & Active Learning in Cambridge. Their team engineers distributed inference engines capable of co-folding and scoring hundreds of millions of virtual compounds while closing the active-learning loop between generative proposals, physics simulations, and wet-lab synthesis.\n\nKEY RESPONSIBILITIES\n• Architect massively parallel GPU/TPU inference pipelines capable of running all-atom structure, docking, and ADMET foundation models across multi-billion-compound virtual libraries.\n• Optimize biomolecular model execution via XLA/TorchInductor compilation, custom triangle-attention kernel fusion, dynamic sequence-length bucketing, and BF16/FP8 precision tuning.\n• Build closed-loop Bayesian optimization and active-learning orchestration engines that balance potency, selectivity, synthetic tractability, and pharmacokinetic multi-parameter objectives.\n• Engineer high-throughput cheminformatics preprocessing, MSA/template retrieval caches, and 3D conformer storage layers sustaining tens of thousands of molecular evaluations per second.\n• Partner with research scientists to distill computationally expensive ensemble structure predictors into rapid student surrogates for early-stage chemical space filtering.\n• Establish deterministic reproducibility, provenance tracking, and uncertainty-calibration dashboards consumed directly by therapeutic project leads.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, C++/Rust, XLA/Triton, Ray, Kubernetes, RDKit, Parquet/Arrow, large-scale cloud GPU/TPU clusters. Workplace: Hybrid in Cambridge, UK. Compensation: £110,000 – £155,000 base salary plus equity/bonus incentives, private medical cover, generous pension, and UK visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of ML engineering, high-performance scientific computing, or inference infrastructure experience.\n• Deep production proficiency in JAX or PyTorch, including model compilation (XLA, TorchCompile, TensorRT), GPU memory optimization, and distributed batch orchestration.\n• Experience designing large-scale distributed pipelines using Ray, Kubernetes, or HPC schedulers processing hundreds of millions of records.\n• Strong software craftsmanship in Python (and C++ or Rust for compute-critical cheminformatics/featurization routines).\n• Working knowledge of Bayesian optimization, active learning, uncertainty quantification, or multi-objective search.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience with cheminformatics toolkits (RDKit), protein MSA search pipelines (MMseqs2/HHblits), or 3D molecular representations.\n• Hands-on experience optimizing pair-representation / axial / triangle attention kernels used in biomolecular transformers.\n• Background scaling computational pipelines within an AI biotech, pharmaceutical discovery, or frontier research lab environment.",
      "updated_at": "2026-11-15T17:33:03.961Z",
      "created_at": "2026-11-15T16:53:16.666Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-engineer-ultra-scale-molecular-inference",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-engineer-ultra-scale-molecular-inference.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-ultra-scale-molecular-inference/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-ultra-scale-molecular-inference/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "d0cc0a50-09dd-44a9-89f8-dd91e4b2796d",
      "slug": "senior-research-scientist-3d-biomolecular",
      "title": "Senior Research Scientist, 3D Biomolecular Foundation Models",
      "department": "AI Research",
      "department_code": "RES",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 135000,
        "max": 190000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£135k – £190k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Scientist, 3D Biomolecular Foundation Models (AI Research) — London, UK · Hybrid · £135k – £190k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an AI-first digital biology and rational drug discovery powerhouse headquartered in London, backed by multi-billion-dollar pharmaceutical partnerships and dedicated frontier compute. Their interdisciplinary research teams build unified all-atom biomolecular foundation models that predict 3D structures, conformational ensembles, and binding affinities across proteins, nucleic acids, covalent ligands, and macrocycles.\n\nThey have retained Talent247 to appoint a Senior Research Scientist in 3D Biomolecular Foundation Models in London. Working closely with medicinal chemists, structural biologists, and ML systems leads, the appointed scientist will pioneer generative 3D structure-and-affinity architectures that directly design clinical-grade therapeutic candidates de novo.\n\nKEY RESPONSIBILITIES\n• Originate and scale SE(3)-equivariant and non-equivariant diffusion, flow-matching, and transformer architectures for joint all-atom biomolecular structure prediction and de novo pocket-conditioned ligand generation.\n• Extend biomolecular foundation models beyond static crystal snapshots to model Boltzmann conformational ensembles, cryptic pocket opening, induced-fit dynamics, and water-mediated interactions.\n• Formulate multi-task post-training and physics-informed fine-tuning objectives uniting experimental cryo-EM/X-ray structures, free-energy perturbation (FEP) landscapes, and high-throughput wet-lab assay readouts.\n• Design rigorous out-of-distribution generalization benchmarks across novel protein folds, covalent warheads, protein-protein interfaces, and ternary degrader complexes.\n• Collaborate directly with active therapeutic program teams to deploy generative design loops against previously undruggable oncology and immunology targets.\n• Lead multi-node scaling ablations in JAX/PyTorch and publish foundational advances where aligned with IP strategy.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, RDKit, OpenMM, geometric deep learning, diffusion/flow-matching frameworks, dedicated multi-thousand GPU/TPU supercomputing clusters. Workplace: Hybrid in London, UK. Compensation: £135,000 – £190,000 base salary plus significant equity/bonus participation, comprehensive benefits, and full UK visa and relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computational Biophysics, Computational Chemistry, Bioinformatics, or Computer Science with 3+ years of post-doctoral or industrial AI research impact.\n• Strong publication or patent record in 3D molecular generative modeling, biomolecular structure prediction, geometric deep learning, or deep learning for drug discovery.\n• Expert-level proficiency in JAX or PyTorch, including custom attention/pair-representation kernels and distributed multi-accelerator training.\n• Solid grounding in structural biology, molecular mechanics, thermodynamics of binding, and stereochemical constraints.\n• Demonstrated ability to iterate rapidly between mathematical model formulation and large-scale empirical validation.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience with Riemannian flow matching, SE(3) diffusion, all-atom biomolecular transformers, or free-energy/affinity prediction networks.\n• Track record impacting live small-molecule, macrocycle, or biologics discovery programs alongside medicinal chemists.\n• Familiarity with molecular dynamics (OpenMM/GROMACS), quantum-chemical datasets (QM9/SPICE), or cryo-EM density map modeling.",
      "updated_at": "2026-11-15T17:32:54.430Z",
      "created_at": "2026-11-15T16:53:12.707Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-scientist-3d-biomolecular",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-scientist-3d-biomolecular.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-3d-biomolecular/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-3d-biomolecular/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "417676f7-4ca0-43a7-8e72-9db1ad62ae12",
      "slug": "senior-ml-data-platform-engineer-petabyte-video",
      "title": "Senior ML Data Platform Engineer, Petabyte Video Mining",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 100000,
        "max": 140000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£100k – £140k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Data Platform Engineer, Petabyte Video Mining (MLOps & Platform) — London, UK · Hybrid · £100k – £140k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a flagship London physical-AI leader scaling embodied foundation models across global vehicle fleets. Because uniform random road driving is overwhelmingly repetitive, their model breakthroughs depend on an intelligent petabyte-scale data engine that continuously mines rare, high-surprise behavioral interactions from hundreds of millions of driven miles.\n\nThey have retained Talent247 to hire a Senior ML Data Platform Engineer for their Petabyte Video Mining team in London. Their group builds multimodal embedding indices, active-learning curation pipelines, and streaming columnar storage engines that turn raw fleet video into high-signal pre-training and post-training mixtures.\n\nKEY RESPONSIBILITIES\n• Architect distributed multimodal ingestion and curation pipelines processing tens of terabytes of daily multi-camera video, kinematics, and audio logs across hybrid cloud and on-prem clusters.\n• Build high-throughput semantic and spatio-temporal retrieval engines over billions of video clips using vision-language embeddings, Lance/Parquet columnar formats, and approximate nearest-neighbor indices.\n• Partner with foundation model researchers to operationalize active-learning and uncertainty-sampling loops that automatically surface high-loss or safety-critical driving episodes from fleet uploads.\n• Engineer zero-copy GPU video transcoding, hardware-accelerated decoding (NVDEC), and deterministic shard shuffling layers feeding multi-thousand-GPU training jobs without IO starvation.\n• Design dataset lineage, deduplication, and mixture-weighting frameworks so scientists can declaratively compose reproducible petabyte training recipes in minutes.\n• Drive storage tiering and compute cost efficiency across object storage, NVMe caches, and spot GPU/CPU fleets.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, Rust, Ray Data, Apache Spark, Lance / Apache Arrow / Parquet, PyTorch, NVDEC/FFmpeg, Kubernetes, AWS/GCP & bare-metal object storage. Workplace: Hybrid in London, UK. Compensation: £100,000 – £140,000 base salary plus equity ownership, comprehensive UK benefits, and Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of data platform, ML infrastructure, or distributed systems engineering experience operating at multi-petabyte scale.\n• Production mastery of distributed compute frameworks such as Ray Data, Apache Spark, or Dask in Python (plus Rust, Go, or C++ for IO-critical components).\n• Hands-on experience with modern multimodal/columnar storage formats (Lance, Parquet, Arrow, WebDataset, Mosaic Streaming) and cloud object-store throughput tuning.\n• Practical familiarity with running batch GPU inference (CLIP/VLM embeddings, auto-labeling, deduplication) across millions of images or video segments.\n• Strong systems fundamentals in networking, concurrent IO, distributed caching, and Kubernetes job orchestration.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience building data curation engines for autonomous driving, robotics, or frontier video/multimodal foundation models.\n• Experience with hardware-accelerated video codecs (H.265/AV1, NVDEC/PyNvVideoCodec) and random-access frame seeking at training time.\n• Contributions to open-source data infrastructure ecosystems (Ray, LanceDB, Arrow, DataFusion, Polars).",
      "updated_at": "2026-11-15T17:32:44.754Z",
      "created_at": "2026-11-15T16:53:09.419Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-data-platform-engineer-petabyte-video",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-data-platform-engineer-petabyte-video.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-data-platform-engineer-petabyte-video/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-data-platform-engineer-petabyte-video/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "dcb2a061-8a6b-40c9-89b8-78898600cd76",
      "slug": "senior-embedded-ai-inference-engineer-cuda",
      "title": "Senior Embedded AI Inference Engineer (CUDA / TensorRT)",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "London, UK (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 110000,
        "max": 150000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£110k – £150k",
        "subtitle": "annual package"
      },
      "summary": "Senior Embedded AI Inference Engineer (CUDA / TensorRT) (AI Infrastructure) — London, UK · On-site · £110k – £150k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier London embodied-AI company deploying end-to-end multimodal foundation models onto commercial and consumer vehicles worldwide. Running billion-parameter vision-language-action (VLA) transformers deterministically inside tight automotive thermal and millisecond latency envelopes is one of their core engineering differentiators.\n\nThey have partnered with Talent247 to appoint a Senior Embedded AI Inference Engineer (CUDA / TensorRT) onsite in London. Sitting between foundation model architects and vehicle platform engineers, the successful candidate will own custom kernel development, low-bit quantization, and deterministic onboard execution across next-generation automotive SoCs.\n\nKEY RESPONSIBILITIES\n• Architect ultra-low-latency C++20, CUDA, and TensorRT / TensorRT-LLM inference pipelines executing multimodal vision-language-action models on automotive-grade NVIDIA Orin and Thor SoCs.\n• Write custom CUDA/PTX and CUTLASS kernels for fused multi-camera patch tokenization, temporal cross-attention, paged KV-caching, and flow-matching action decoding.\n• Co-design INT8, FP8, and INT4/AWQ quantization-aware training and post-training calibration workflows with ML scientists to achieve 3–4x throughput gains with zero safety regression.\n• Eliminate tail-latency jitter through CUDA Graph capture, zero-copy DMA sensor ingestion, deterministic memory arenas, and real-time Linux/RTOS thread scheduling.\n• Build automated hardware-in-the-loop (HIL) profiling suites tracking layer-by-layer roof-line utilization, SRAM/DRAM bandwidth, power draw, and numerical parity against PyTorch reference checkpoints.\n• Support global OEM partner integrations by abstracting hardware-specific graph compilers and safety-island watchdogs behind clean deployment interfaces.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Modern C++20, CUDA, PTX, CUTLASS, TensorRT / TensorRT-LLM, ONNX/MLIR, PyTorch, Nsight Systems & Compute, Linux RT / QNX, automotive GPU SoCs. Workplace: Onsite in London, UK (direct access to vehicle hardware-in-the-loop benches and test fleet). Compensation: £110,000 – £150,000 base salary plus generous equity, private healthcare, and full relocation/visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems and inference engineering experience in modern C++ (C++17/20) and CUDA on NVIDIA GPUs.\n• Deep production mastery of TensorRT or TensorRT-LLM, including custom plugin authoring, graph fusion debugging, and mixed-precision calibration (FP16/FP8/INT8/INT4).\n• Demonstrated ability to profile GPU hardware bottlenecks (warp occupancy, shared memory bank conflicts, tensor core utilization) using Nsight Compute and Nsight Systems.\n• Strong understanding of transformer, ViT, and temporal sequence model architectures and KV-cache memory mechanics.\n• Experience building deterministic, safety-conscious real-time software on embedded Linux or RTOS platforms.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct hands-on experience with NVIDIA DRIVE Orin/Thor, Jetson AGX, or similar automotive/robotics edge accelerators.\n• Familiarity with MLIR/TVM/XLA compiler passes, quantization-aware training (QAT) in PyTorch, or speculative decoding on edge hardware.\n• Exposure to ISO 26262 functional safety concepts, zero-copy camera ISP pipelines (GMSL/NvMedia), or robotics middleware.",
      "updated_at": "2026-11-15T17:32:35.979Z",
      "created_at": "2026-11-15T16:53:05.724Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-embedded-ai-inference-engineer-cuda",
        "markdown": "https://jobs.talent247.co/jobs/senior-embedded-ai-inference-engineer-cuda.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-embedded-ai-inference-engineer-cuda/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-embedded-ai-inference-engineer-cuda/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "51fb1638-8c2b-405e-a3be-6987fcd3a61a",
      "slug": "senior-ml-engineer-neural-world-simulation-video",
      "title": "Senior ML Engineer, Neural World Simulation & Video Diffusion",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 115000,
        "max": 160000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£115k – £160k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Engineer, Neural World Simulation & Video Diffusion (Robotics & Embodied AI) — London, UK · Hybrid · £115k – £160k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a category-defining embodied AI company in London building mapless, end-to-end foundation models for autonomous vehicles. To evaluate and train safety-critical driving policies beyond recorded road logs, their Simulation & World Modeling group builds photorealistic, reactive neural simulators capable of synthesizing counterfactual multi-camera traffic scenarios on demand.\n\nThey have engaged Talent247 to hire a Senior ML Engineer specializing in Neural World Simulation & Video Diffusion in London. Their team combines 3D Gaussian Splatting, neural radiance fields, and multi-view latent video diffusion to turn petabytes of fleet driving captures into controllable, closed-loop training environments.\n\nKEY RESPONSIBILITIES\n• Train and scale multi-camera, action-conditioned latent video diffusion and autoregressive world models that maintain geometric cross-view consistency under novel ego-vehicle trajectories.\n• Develop hybrid 3D Gaussian Splatting (3DGS) and generative diffusion pipelines capable of reconstructing dynamic urban scenes and editing weather, lighting, and reactive traffic actors.\n• Optimize inference latency of neural sensor simulators via step distillation, FP8/BF16 mixed precision, and custom Triton/CUDA attention kernels so RL policies can roll out at scale.\n• Engineer automated scene-reconstruction pipelines that process thousands of hours of raw multi-sensor fleet logs with calibrated extrinsics, rolling-shutter compensation, and dynamic object tracks.\n• Design perceptual realism, 3D epipolar consistency, and domain-gap metrics verifying that simulated rollouts correlate tightly with real-road policy behavior.\n• Collaborate closely with RL post-training scientists to synthesize adversarial edge-case scenarios (cut-ins, occlusions, vulnerable road users) for targeted policy hardening.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, CUDA/Triton, Diffusers/custom flow-matching frameworks, gsplat / 3DGS / NeRF, Ray, large-scale H100 GPU clusters. Workplace: Hybrid in London, UK. Compensation: £115,000 – £160,000 base salary plus meaningful equity, comprehensive private health coverage, and UK visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of research engineering or ML engineering experience in 3D computer vision, neural rendering, or generative video modeling (or PhD with strong engineering output).\n• Deep hands-on experience training latent video diffusion, flow-matching, DiT (Diffusion Transformer), or 3D Gaussian Splatting models in PyTorch.\n• Strong grasp of multi-view camera geometry, projective transformations, depth estimation, and spatiotemporal attention architectures.\n• Proven experience scaling distributed multi-GPU training runs and diagnosing throughput, memory, and numerical stability bottlenecks.\n• Clean, modular Python/C++ engineering practices suitable for production-grade simulation infrastructure.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience with autonomous driving sensor suites (surround-view cameras, LiDAR, radar) or closed-loop driving simulators.\n• Expertise in diffusion distillation (LCM, DMD, consistency models) or causal streaming video generation.\n• First-author publications or major open-source contributions in CVPR, ICCV, ECCV, SIGGRAPH, or NeurIPS.",
      "updated_at": "2026-11-15T17:32:26.141Z",
      "created_at": "2026-11-15T16:53:03.145Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-engineer-neural-world-simulation-video",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-engineer-neural-world-simulation-video.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-neural-world-simulation-video/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-neural-world-simulation-video/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "1a959b61-f283-4b04-b9fd-e287ce35f5f0",
      "slug": "principal-ml-scientist-embodied-autonomous-driving",
      "title": "Principal ML Scientist, Embodied Autonomous Driving Models",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 140000,
        "max": 190000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£140k – £190k",
        "subtitle": "annual package"
      },
      "summary": "Principal ML Scientist, Embodied Autonomous Driving Models (Robotics & Embodied AI) — London, UK · Hybrid · £140k – £190k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a billion-dollar-backed London physical-AI pioneer that has redefined autonomous mobility through end-to-end embodied foundation models. Rather than relying on brittle HD maps and hand-coded rules, their mapless vision-language-action (VLA) driving intelligence generalizes zero-shot across complex urban environments in Europe, North America, and Asia in partnership with global automotive OEMs.\n\nThey have retained Talent247 to identify a Principal ML Scientist for their Embodied Autonomous Driving Models organization in London. Serving as a technical pillar across pre-training and closed-loop post-training, the successful candidate will shape the next generation of unified multimodal driving architectures that fuse surround-view video, radar, language reasoning, and continuous trajectory generation.\n\nKEY RESPONSIBILITIES\n• Define the multi-quarter architectural roadmap for multi-billion-parameter embodied driving foundation models uniting causal spatiotemporal perception, chain-of-thought scene reasoning, and continuous control.\n• Pioneer closed-loop reinforcement learning and imitation learning post-training paradigms that eliminate causal confusion, covariate shift, and rare long-tail safety failures.\n• Architect cross-embodiment sensor-abstraction layers allowing a single pretrained driving backbone to transfer seamlessly across passenger vehicles, commercial fleets, and diverse camera/radar rigs.\n• Co-design model distillation and quantization-aware architectures alongside onboard inference teams to meet strict real-time automotive compute and latency budgets.\n• Establish rigorous closed-loop neural-simulation benchmarks and safety-critical behavioral evaluation gates prior to public-road fleet deployment.\n• Provide technical mentorship to senior scientists and ML engineers while partnering directly with executive research leadership on compute allocation and data strategy.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, PyTorch, FSDP/Megatron-Core, Triton/CUDA, multi-thousand H100/B200 GPU training clusters, petabyte fleet driving datasets, neural closed-loop simulators. Workplace: Hybrid in London, UK. Compensation: £140,000 – £190,000 base salary plus high-upside late-stage equity, private healthcare, and full UK visa/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD or equivalent Principal/Staff industrial research tenure (7+ years) in Machine Learning, Robotics, Computer Vision, or Autonomous Systems.\n• Proven track record architecting and scaling large multimodal foundation models, VLA policies, or end-to-end autonomous driving networks on multi-node GPU clusters.\n• Deep expertise in imitation learning, offline/online reinforcement learning, diffusion/flow-matching action heads, and sequence modeling in PyTorch.\n• Strong intuition for data scaling laws, curriculum curation across petabyte-scale video logs, and closed-loop vs. open-loop metric divergence.\n• Track record setting technical direction across multi-team research-to-production initiatives.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience deploying learned end-to-end planners or perception-to-control models onto real-world test fleets or physical robotic platforms.\n• Publications at CVPR, RSS, CoRL, ICRA, NeurIPS, or ICML in embodied AI, world models, or autonomous driving.\n• Experience distilling frontier-scale teacher models into low-latency student networks targeting embedded automotive SoCs.",
      "updated_at": "2026-11-15T17:32:16.220Z",
      "created_at": "2026-11-15T16:52:59.095Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/principal-ml-scientist-embodied-autonomous-driving",
        "markdown": "https://jobs.talent247.co/jobs/principal-ml-scientist-embodied-autonomous-driving.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/principal-ml-scientist-embodied-autonomous-driving/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/principal-ml-scientist-embodied-autonomous-driving/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "1ced433d-4943-4102-bf8b-443e60e08e0b",
      "slug": "research-scientist-ai-for-mathematics-physical",
      "title": "Research Scientist, AI for Mathematics & Physical Sciences",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Cambridge, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 130000,
        "max": 185000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£130k – £185k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, AI for Mathematics & Physical Sciences (AI Research) — Cambridge, UK · Hybrid · £130k – £185k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier frontier AI research laboratory expanding their interdisciplinary scientific discovery hub in Cambridge, UK. Their AI for Science & Formal Reasoning division applies foundation models, geometric deep learning, and reinforcement learning to solve grand challenges in pure mathematics, quantum chemistry, materials discovery, and plasma physics.\n\nThey have retained Talent247 to appoint a Research Scientist in AI for Mathematics & Physical Sciences in Cambridge. Collaborating with Fields-caliber mathematicians, theoretical physicists, and ML systems engineers, the appointed scientist will develop neuro-symbolic theorem provers and physically constrained neural surrogates that discover verifiable scientific insights.\n\nKEY RESPONSIBILITIES\n• Formulate novel machine learning architectures and reinforcement learning search methods for formal mathematical reasoning and automated theorem proving in interactive proof assistants such as Lean 4.\n• Develop equivariant geometric neural networks, neural operators, and score-based generative models for quantum many-body systems, crystalline materials, and partial differential equations (PDEs).\n• Build closed-loop synthetic conjecture generation and formal verification curricula that overcome human-proof data scarcity through self-play and verifier feedback.\n• Collaborate with academic and national-laboratory partners to validate model predictions against ab-initio DFT simulations, fusion plasma diagnostics, or formal proof libraries.\n• Scale training and search inference experiments across multi-node GPU and custom accelerator clusters using JAX.\n• Publish seminal results in leading machine learning and natural science journals while open-sourcing benchmarks that advance the scientific AI ecosystem.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, Lean 4 / Isabelle, equivariant GNNs (e3nn), differentiable physics solvers, multi-node HPC/accelerator clusters. Workplace: Hybrid in Cambridge, UK. Compensation: £130,000 – £185,000 base salary plus equity/annual bonus, comprehensive benefits, and full UK Global Talent / Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Mathematics, Theoretical Physics, Computational Chemistry, or Computer Science with a strong frontier research record.\n• First-author publications at top ML conferences (NeurIPS, ICML, ICLR) or leading scientific journals (Nature, Science, PRL, JACS, Communications in Mathematical Physics).\n• Deep proficiency in JAX or PyTorch, differentiable programming, and scaling neural architectures on distributed compute.\n• Rigorous mathematical foundations in at least one target domain: formal logic/type theory, differential geometry, quantum mechanics, or numerical PDEs.\n• Demonstrated ability to bridge abstract domain symmetries and constraints with clean, scalable deep learning implementations.\n\nPREFERRED / BONUS EXPERIENCE\n• Hands-on experience interfacing LLMs or tree-search RL agents with Lean 4 (Mathlib), Coq, or Isabelle proof kernels.\n• Prior work on SE(3)/E(3)-equivariant neural networks, neural operators (FNO), flow matching on manifolds, or DFT/molecular dynamics surrogates.\n• Olympiad mathematics/physics background or contributions to major open formalization or scientific ML repositories.",
      "updated_at": "2026-11-15T17:32:05.601Z",
      "created_at": "2026-11-15T16:52:55.428Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "5d52ae7f-e5c7-4002-a267-94b312835920",
      "slug": "research-engineer-multimodal-safety-post-training",
      "title": "Research Engineer, Multimodal Safety & Post-Training",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 125000,
        "max": 175000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£125k – £175k",
        "subtitle": "annual package"
      },
      "summary": "Research Engineer, Multimodal Safety & Post-Training (Machine Learning) — London, UK · Hybrid · £125k – £175k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an elite frontier AI research lab in Central London developing flagship multimodal reasoning models deployed to hundreds of millions of users and enterprise developers globally. Their Safety & Alignment Engineering organization ensures that rapidly advancing capabilities in autonomous tool use, multimodal perception, and code synthesis remain robustly aligned, refusal-calibrated, and resistant to adversarial exploitation.\n\nThey have engaged Talent247 to recruit a Research Engineer specializing in Multimodal Safety & Post-Training in London. Their team sits directly in the critical path of flagship model launches, building automated red-teaming agents, constitutional preference pipelines, and post-training interventions across text, audio, image, and video modalities.\n\nKEY RESPONSIBILITIES\n• Design and execute post-training alignment interventions—spanning SFT, DPO, RLAIF, and online RL—that enforce nuanced safety specifications without degrading helpfulness or false-refusal calibration.\n• Build autonomous multi-turn adversarial red-teaming agents that probe frontier multimodal models for cross-modal prompt injection, steganographic jailbreaks, cyber-offense uplift, and agentic sandbagging.\n• Train high-precision multimodal reward models and generative safety classifiers capable of reasoning over interleaved visual, audio, and code tool-execution traces.\n• Construct scalable synthetic preference data engines that target rare, high-severity edge cases and ambiguous dual-use scientific queries.\n• Partner with pre-training, post-training, and policy governance leads to define automated launch-readiness safety gates and quantitative risk scorecards.\n• Analyze mechanistic failure modes across training checkpoints to diagnose reward hacking, sycophancy, and deceptive alignment behaviors.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, distributed RLHF/RLAIF frameworks, sandboxed agent evaluation harnesses, large-scale accelerator clusters. Workplace: Hybrid in London, UK. Compensation: £125,000 – £175,000 base salary plus equity/bonus incentives, private medical insurance, pension match, and UK visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• BS, MS, or PhD in Computer Science, Machine Learning, or a related quantitative discipline with 4+ years of hands-on deep learning and LLM post-training experience.\n• Proven track record implementing SFT, RLHF, RLAIF, DPO, GRPO, or PPO pipelines on large language or vision-language models.\n• Strong systems and ML engineering skills in JAX or PyTorch, including distributed evaluation orchestration and dataset quality diagnostics.\n• Practical experience designing adversarial benchmarks, red-teaming harnesses, or calibrated LLM-as-a-judge evaluation pipelines.\n• Clear analytical mindset capable of translating high-level safety principles into measurable optimization objectives.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience securing multimodal (audio/video/vision) or tool-using computer-use agents against indirect prompt injection and privilege escalation.\n• Publications or open-source contributions in AI alignment, adversarial robustness, representation engineering, or calibration.\n• Familiarity with activation steering, linear probes, or representation-level circuit monitoring during post-training.",
      "updated_at": "2026-11-15T17:31:56.575Z",
      "created_at": "2026-11-15T16:52:52.177Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-engineer-multimodal-safety-post-training",
        "markdown": "https://jobs.talent247.co/jobs/research-engineer-multimodal-safety-post-training.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-engineer-multimodal-safety-post-training/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-engineer-multimodal-safety-post-training/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "6e6ad41d-56bc-47b7-a7a8-f7f9aa2d7668",
      "slug": "senior-research-engineer-generative-world-models",
      "title": "Senior Research Engineer, Generative World Models",
      "department": "AI Research",
      "department_code": "RES",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 140000,
        "max": 195000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£140k – £195k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Engineer, Generative World Models (AI Research) — London, UK · Hybrid · £140k – £195k. THE CLIENT & MANDATE Our client is a globally…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally recognized frontier AI research organization headquartered in London, uniting fundamental machine learning scientists and large-scale systems engineers to solve general-purpose intelligence. Their generative world-modeling division builds real-time, interactive spatiotemporal foundation models that simulate physical dynamics, 3D geometry, and causal agent interactions from raw multimodal streams.\n\nThey have partnered with Talent247 to hire a Senior Research Engineer for their Generative World Models team in London. Working directly alongside principal research scientists, the post-holder will architect causal video tokenizers, long-horizon memory mechanisms, and high-throughput multi-node training pipelines that bring interactive world simulators to real-time frame rates.\n\nKEY RESPONSIBILITIES\n• Co-design and scale autoregressive, masked-generative, and flow-matching world models capable of maintaining 3D spatial and physical consistency over multi-minute interactive rollouts.\n• Engineer compact spatiotemporal latent tokenizers and continuous-discrete variational autoencoders optimized for temporal compression and high visual fidelity.\n• Build deterministic, fault-tolerant petabyte-scale video and multimodal action-conditioned data pipelines feeding multi-slice accelerator training jobs.\n• Optimize long-context spatiotemporal attention kernels, KV-cache eviction strategies, and speculative frame decoding to achieve interactive sub-50ms step latencies.\n• Construct automated physics-consistency, object-permanence, and counterfactual action-controllability evaluation suites.\n• Drive end-to-end experimental velocity from clean single-node ablations through full-scale foundation model training runs.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, PyTorch, XLA, custom CUDA/Pallas kernels, distributed petabyte video loaders, multi-pod TPU/GPU supercomputing clusters. Workplace: Hybrid in London, UK. Compensation: £140,000 – £195,000 base salary plus competitive equity/bonus package, comprehensive UK benefits, and full visa and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• MS or PhD in Computer Science, Machine Learning, Computer Vision, or equivalent Senior-level research engineering experience in generative modeling.\n• Demonstrated production or frontier-lab experience training large-scale video diffusion, flow-matching, or autoregressive multimodal foundation models.\n• Mastery of JAX or PyTorch across multi-host data, tensor, sequence, and pipeline parallelism regimes.\n• Strong software engineering discipline in Python (and C++ where performance-critical), including reproducible configuration, profiling, and automated testing.\n• Deep familiarity with 3D spatial representations, camera geometry, optical flow, or latent video compression architectures.\n\nPREFERRED / BONUS EXPERIENCE\n• Published research at CVPR, ICCV, ECCV, NeurIPS, ICML, or ICLR on video generation, world models, neural rendering, or embodied simulation.\n• Hands-on experience distilling multi-step diffusion models into single-step or few-step real-time causal generators (consistency distillation, adversarial diffusion distillation).\n• Familiarity with coupling learned world models to downstream reinforcement learning or planning agents.",
      "updated_at": "2026-11-15T17:31:45.832Z",
      "created_at": "2026-11-15T16:52:48.716Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-engineer-generative-world-models",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-engineer-generative-world-models.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-generative-world-models/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-generative-world-models/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "2c05625d-9aa1-49fb-bbe2-a5dd0333a0f2",
      "slug": "staff-research-scientist-reinforcement-learning",
      "title": "Staff Research Scientist, Reinforcement Learning & Planning",
      "department": "AI Research",
      "department_code": "RES",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 165000,
        "max": 235000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£165k – £235k",
        "subtitle": "annual package"
      },
      "summary": "Staff Research Scientist, Reinforcement Learning & Planning (AI Research) — London, UK · Hybrid · £165k – £235k. THE CLIENT & MANDATE Our client is one…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of the world's foremost frontier artificial intelligence research laboratories, operating from a flagship research campus in London's King's Cross knowledge quarter. Their reinforcement learning, search, and sequential decision-making groups have pioneered foundational breakthroughs in self-play, planning under uncertainty, and test-time compute scaling across complex reasoning and scientific domains.\n\nThey have retained Talent247 to appoint a Staff Research Scientist in Reinforcement Learning & Planning in London. Sitting within their core frontier reasoning pillar, the successful candidate will set the research agenda for next-generation RL algorithms that unify learned world representations, Monte Carlo tree search, and large-scale policy optimization across multi-thousand-accelerator clusters.\n\nKEY RESPONSIBILITIES\n• Originate and validate novel reinforcement learning and test-time search algorithms capable of multi-step planning, backtracking, and self-verification across open-ended reasoning, coding, and mathematical domains.\n• Architect scalable distributed actor-learner and asynchronous rollout topologies in JAX and Python across multi-slice GPU and custom accelerator pods.\n• Design dense process-reward formulations, credit-assignment mechanisms, and off-policy value estimators that remain stable across long-horizon trajectories exceeding tens of thousands of tokens.\n• Investigate the scaling laws governing pre-training priors, synthetic self-play data generation, and inference-time search compute budgets.\n• Partner closely with distributed systems engineers and post-training leads to transition high-signal algorithmic prototypes directly into flagship frontier model training runs.\n• Mentor fellow research scientists and engineers while contributing to internal technical reports and top-tier peer-reviewed publications.\n\nTECHNICAL ENVIRONMENT & OFFERING\nStack: Python, JAX, Flax/Haiku, PyTorch, XLA, distributed RL actor-learner frameworks, multi-slice custom ASIC and GPU clusters. Workplace: Hybrid in London (King's Cross), UK. Compensation: £165,000 – £235,000 base salary plus substantial equity/bonus participation, comprehensive private healthcare, generous pension contributions, and full UK Global Talent / Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computer Science, Control Theory, Mathematics, or equivalent Staff-level frontier research track record in reinforcement learning and planning.\n• Strong first-author publication record at venues such as NeurIPS, ICML, ICLR, RLDM, or JMLR covering RL, search/planning, bandits, or sequential decision-making.\n• Deep hands-on proficiency in JAX or PyTorch, including custom vectorized loss implementations and multi-node distributed training diagnostics.\n• Proven experience scaling reinforcement learning or post-training algorithms on large transformer or state-space architectures.\n• Ability to formulate crisp empirical ablations that isolate algorithmic signal from compute or hyperparameter confounding.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior breakthroughs in Monte Carlo Tree Search (MCTS), alpha-beta/guided search, self-play curricula, or RL with verifiable rewards (RLVR).\n• Experience profiling XLA compilation graphs, collective communications, and memory-bound rollout bottlenecks on multi-slice accelerator topologies.\n• Track record of technically leading multi-researcher initiatives within an industrial frontier AI laboratory.",
      "updated_at": "2026-11-15T17:31:37.129Z",
      "created_at": "2026-11-15T16:52:46.609Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-research-scientist-reinforcement-learning",
        "markdown": "https://jobs.talent247.co/jobs/staff-research-scientist-reinforcement-learning.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-research-scientist-reinforcement-learning/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-research-scientist-reinforcement-learning/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "9c6419ba-7274-4580-a072-1c32110b2d49",
      "slug": "staff-scientific-ml-platform-compute-engineer",
      "title": "Staff Scientific ML Platform & Compute Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 120000,
        "max": 165000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£120k – £165k",
        "subtitle": "annual package"
      },
      "summary": "Staff Scientific ML Platform & Compute Engineer (MLOps & Platform) — London, UK · Hybrid · £120k – £165k. THE CLIENT & MANDATE Our client is an AI-first…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an AI-first drug discovery powerhouse headquartered in London (with a second UK research hub in Cambridge), spun out of one of the world's most accomplished frontier AI research organizations and backed by hundreds of millions in capital alongside multi-billion-dollar biopharma partnerships. They are building a unified, predictive and generative 3D biomolecular foundation model platform capable of modeling how proteins, nucleic acids, and small-molecule therapeutics interact at atomic resolution.\n\nThey have retained Talent247 to appoint a Staff Scientific ML Platform & Compute Engineer in London. Sitting as a senior technical architect bridging ML research scientists, computational chemists, and large-scale accelerator infrastructure, you will build the core training/inference compute platform, reproducible structural-biology data engine, and interactive 3D model-serving surfaces that power live therapeutic discovery programmes.\n\nKEY RESPONSIBILITIES\n• Set technical direction and own the core ML engineering platform supporting multi-slice TPU and GPU pre-training, fine-tuning, and massive-scale virtual screening runs in JAX and PyTorch.\n• Architect petabyte-scale structural, genomic, and chemical data pipelines—managing crystallography/cryo-EM structures, multiple-sequence alignments (MSAs), molecular dynamics trajectories, and proprietary wet-lab assay results with strict lineage and zero train-test leakage.\n• Build low-latency, high-availability interactive model-serving APIs so medicinal chemists and structural biologists can co-fold, dock, and score candidate molecule edits in real time inside 3D molecular design workspaces.\n• Optimize distributed compilation (XLA), host-to-device input bottlenecks, checkpointing, and cluster job scheduling to maximize researcher iteration velocity across hundreds of concurrent experiments.\n• Enforce bank-grade security, auditability, and multi-partner data isolation protecting high-value pharmaceutical collaboration IP.\n• Mentor senior ML and platform engineers, establishing testing, profiling, and engineering excellence across the shared scientific codebase.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack centers on Python, JAX/XLA, PyTorch, C++/Rust high-speed structural parsers, Kubernetes on GCP, Ray/Apache Beam, BigQuery/Parquet, and massive TPU/GPU compute allocations. Working hybrid in central London (King's Cross ecosystem), our client offers £120,000–£165,000 GBP base salary plus competitive equity/long-term incentives, premier pension and private medical benefits, and full visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of software, ML platform, or distributed infrastructure engineering experience, including proven Staff-level impact supporting frontier ML research or scientific computing organizations.\n• Deep production mastery of Python and cloud-native distributed systems on GCP or AWS (Kubernetes, high-throughput object storage, distributed batch/streaming compute, Infrastructure-as-Code).\n• Direct hands-on experience optimizing and operating large-scale JAX/XLA or PyTorch training and inference workloads on multi-node GPU or TPU clusters.\n• Strong engineering instincts for data-pipeline determinism, schema evolution, reproducible experiment tracking, and developer productivity tooling.\n• Collaborative, low-ego leadership style able to bridge deep systems engineers, ML theorists, and drug-discovery scientists.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior exposure to structural biology, cheminformatics, or bioinformatics file formats and toolchains (PDB/mmCIF, RDKit, OpenMM, HHblits/MMseqs2 MSA pipelines).\n• Proficiency in Rust or C++ for accelerating domain-specific geometric or molecular preprocessing bottlenecks.",
      "updated_at": "2026-11-15T17:28:47.633Z",
      "created_at": "2026-11-15T16:53:19.844Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-scientific-ml-platform-compute-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-scientific-ml-platform-compute-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-scientific-ml-platform-compute-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-scientific-ml-platform-compute-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "bf64dc11-f987-4c3f-891b-c851e6cfd6ca",
      "slug": "research-scientist-generative-video-neural-avatars",
      "title": "Research Scientist, Generative Video & Neural Avatars",
      "department": "AI Research",
      "department_code": "RES",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 120000,
        "max": 170000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£120k – £170k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Generative Video & Neural Avatars (AI Research) — London, UK · Hybrid · £120k – £170k. THE CLIENT & MANDATE Our client is a $4B…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a $4B generative video pioneer headquartered in London, backed by tier-one global venture capital and strategic AI chipmaker investors, whose synthetic media platform is used by over 90% of the Fortune 100. Within their 40+ person London R&D organization, their Interactive Avatars & Generative Video Research group is pushing the frontier of photorealistic digital humans that move, gesture, emote, listen, and converse indistinguishable from live studio footage.\n\nThey have engaged Talent247 to recruit a Research Scientist, Generative Video & Neural Avatars in London. In this high-ownership research role, you will invent and scale next-generation audio- and intent-conditioned video diffusion and flow-matching models capable of generating expressive full-body human motion, micro-expressions, contextual listener reactions, and rock-solid identity preservation over long interactive horizons.\n\nKEY RESPONSIBILITIES\n• Conceptualize, prototype, and pre-train large spatiotemporal diffusion and flow-matching transformer architectures conditioned on streaming speech audio, text intent, 3D pose/camera controls, and user interaction cues.\n• Solve core scientific challenges in long-horizon interactive avatar streaming: preventing temporal drift, maintaining sub-frame lip-sync accuracy across 140+ languages, and synthesizing natural listener back-channeling (nods, gaze shifts, smiles).\n• Explore 3D-aware priors, implicit human representations, and world-model physics so digital presenters interact coherently with scenes, lighting, and objects.\n• Lead post-training initiatives—including DPO/RLHF for human perceptual preference, identity/motion disentanglement, and step distillation—to bring research breakthroughs to commercial shipping quality.\n• Collaborate closely with capture-studio, multimodal data-curation, and GPU inference teams to co-design high-value training datasets and real-time serving architectures.\n• Publish selective, high-impact work and help shape the multi-year research roadmap for interactive multimodal agents.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers operate on dedicated multi-node H100/B200 GPU clusters with proprietary studio multi-camera capture datasets and a modern PyTorch distributed training stack. Based hybrid in central London, our client offers £120,000–£170,000 GBP base salary plus substantial Series E equity, comprehensive UK health and family benefits, and full visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Computer Vision, Machine Learning, or Computer Graphics (or equivalent industrial research track record) with first-author publications at CVPR, ICCV, ECCV, SIGGRAPH, NeurIPS, ICML, or ICLR.\n• Deep theoretical and hands-on mastery of modern generative visual modeling: video diffusion transformers (DiTs), flow matching, spatiotemporal VAEs/tokenizers, and multi-modal conditioning.\n• High proficiency in PyTorch, including multi-node distributed training, custom loss design, mixed-precision optimization, and perceptual video evaluation.\n• Track record owning end-to-end research projects from literature synthesis and dataset formulation through large-scale training and ablation studies.\n• Strong communication skills and enthusiasm for seeing frontier research used daily by millions of people.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct research experience in 3D/2D digital human avatars, audio-driven facial/body animation, talking-head synthesis, or interactive world models.\n• Familiarity with causal video streaming, distillation, or joint audio-video generation architectures.",
      "updated_at": "2026-11-15T17:28:47.059Z",
      "created_at": "2026-11-15T16:53:23.520Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-generative-video-neural-avatars",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-generative-video-neural-avatars.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-generative-video-neural-avatars/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-generative-video-neural-avatars/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7a2bfb26-915c-4d9d-848b-508eb346b088",
      "slug": "senior-applied-ai-engineer-multimodal-creative",
      "title": "Senior Applied AI Engineer, Multimodal Creative Agents",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 130000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£95k – £130k",
        "subtitle": "annual package"
      },
      "summary": "Senior Applied AI Engineer, Multimodal Creative Agents (Applied AI) — London, UK · Hybrid · £95k – £130k. THE CLIENT & MANDATE Our client is a $4B-valued…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a $4B-valued generative AI video unicorn headquartered in London, relied upon by over 90% of the Fortune 100 to produce studio-quality enterprise communications, product walkthroughs, and interactive learning experiences across 140+ languages without cameras or studios. Having mastered photorealistic neural avatars, they are now building an autonomous multimodal creative agent layer that converts raw enterprise knowledge—documents, slide decks, screen recordings, and URLs—into complete, multi-scene visual productions.\n\nThey have partnered with Talent247 to hire a Senior Applied AI Engineer, Multimodal Creative Agents in London. Embedded in a product-facing AI engineering squad partnering directly with their 40+ person R&D lab, you will architect the multi-agent planning, layout synthesis, and VLM quality-assurance loops that orchestrate scriptwriting, scene composition, avatar direction, b-roll generation, and multilingual localization end to end.\n\nKEY RESPONSIBILITIES\n• Architect and ship multi-step agentic workflows that ingest messy enterprise PDFs, decks, and screen captures, plan narrative arcs, generate structured scene-graph layouts, and direct avatar gesture/prosody cues automatically.\n• Build closed-loop Vision-Language Model (VLM) critique and self-correction agents that inspect rendered video frames for typography overflow, visual contrast, timing pacing, and brand-kit compliance before presentation to the user.\n• Design low-latency collaborative co-editing experiences in TypeScript/React and Python where human creators and AI agents iterate seamlessly on live scene timelines via natural language and direct manipulation.\n• Construct automated multimodal evaluation benchmarks combining human creative-director rubrics, VLM judges, and production user-publish conversion metrics to guide model selection, prompt compilation, and fine-tuning.\n• Optimize cost, reliability, and streaming responsiveness through intelligent routing across frontier multimodal LLMs and custom distilled internal models.\n• Collaborate closely with research engineers on interactive real-time video-avatar agent APIs.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack combines TypeScript, React, Node.js, Python, PyTorch/vLLM services, multimodal VLMs, WebRTC/WebCodecs media timelines, and Kubernetes on AWS. Working hybrid from central London, our client offers £95,000–£130,000 GBP base salary plus Series E unicorn equity, private healthcare, generous parental and holiday leave, and UK visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of full-stack and applied AI software engineering experience shipping polished, high-scale AI-native web products.\n• High proficiency in TypeScript (backend & complex interactive React frontends) and Python for AI orchestration and evaluation pipelines.\n• Proven production experience building structured multi-agent LLM/VLM workflows, tool-calling state machines, context engineering, and quantitative evaluation suites.\n• Strong product and UX instincts for translating ambiguous creative tasks into reliable, steerable human-AI collaborative interfaces.\n• Clean software architecture discipline across asynchronous media jobs, streaming SSE/WebSockets, and observability.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience with programmatic video/canvas/layout engines (Remotion, WebCodecs, Canvas/WebGL, FFmpeg), design-tool editors, or multimodal vision-language models.\n• Experience fine-tuning smaller task-specific models (LoRA/SFT) to replace slow or expensive frontier-model steps.",
      "updated_at": "2026-11-15T17:28:36.071Z",
      "created_at": "2026-11-15T16:53:30.442Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-multimodal-creative",
        "markdown": "https://jobs.talent247.co/jobs/senior-applied-ai-engineer-multimodal-creative.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-multimodal-creative/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-applied-ai-engineer-multimodal-creative/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8c2c7dd5-1ee4-4940-ad27-bd67d7643320",
      "slug": "senior-audio-ml-research-engineer-speech-to-speech",
      "title": "Senior Audio ML Research Engineer, Speech-to-Speech Models",
      "department": "AI Research",
      "department_code": "RES",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 125000,
        "max": 175000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£125k – £175k",
        "subtitle": "annual package"
      },
      "summary": "Senior Audio ML Research Engineer, Speech-to-Speech Models (AI Research) — London, UK · Hybrid · £125k – £175k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a flagship generative audio research lab and multi-billion-dollar unicorn with a major R&D center in central London, widely credited with defining the state of the art in expressive human-level speech synthesis, voice cloning, and real-time voice interaction. Their next frontier moves beyond cascaded ASR-to-LLM-to-TTS pipelines toward unified, native speech-to-speech foundation models that reason directly over interleaved audio and text tokens—capturing laughter, hesitation, emphasis, back-channeling, and fluid duplex overlap without rigid turn-taking.\n\nThey have retained Talent247 to appoint a Senior Audio ML Research Engineer, Speech-to-Speech Models in London. Working inside a small, high-autonomy R&D team alongside world-class speech and multimodal researchers, you will design, pre-train, and post-train end-to-end conversational audio foundation models shipped to millions of users and developers globally.\n\nKEY RESPONSIBILITIES\n• Propose, implement, and scale novel interleaved audio-text autoregressive, flow-matching, and discrete/continuous neural audio codec architectures for real-time full-duplex speech-to-speech generation.\n• Train ultra-low-frame-rate, semantic-rich neural audio tokenizers balancing acoustic fidelity, speaker identity preservation, paralinguistic expressiveness, and causal streaming latency (<80ms lookahead).\n• Lead post-training alignment (DPO, RLHF, distillation) for spoken conversational models—optimizing instruction following, emotional appropriateness, natural interruption handling, and voice consistency across 30+ minute dialogues.\n• Define new latency-aware and interaction-aware evaluation frameworks measuring duplex naturalness, paralinguistic accuracy, and spoken reasoning retention.\n• Curate and filter hundred-thousand-hour multi-channel conversational audio corpora with rich paralinguistic and speaker-turn annotations.\n• Partner closely with CUDA/inference engineers to ship quantized streaming models onto production GPU fleets.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers work in Python and PyTorch across dedicated multi-node H100/B200 clusters with custom distributed audio data loaders and fast production deployment paths. Based hybrid in London, our client offers £125,000–£175,000 GBP base salary plus high-upside unicorn equity, comprehensive UK benefits, and full Global Talent / Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of ML research and engineering experience in generative speech synthesis (TTS), neural audio codecs, speech language models, or multimodal audio-text modeling.\n• Expert proficiency in PyTorch, distributed multi-node training (FSDP/Megatron), causal streaming transformers, flow matching/diffusion, or RVQ/FSQ discrete tokenizers.\n• Proven track record taking novel generative audio or sequence architectures from initial hypothesis and dataset assembly all the way to shipping-quality checkpoints.\n• Deep intuition for time-series/acoustic modeling, receptive fields, streaming lookahead constraints, and audio perceptual quality tradeoffs.\n• Strong general software engineering discipline within a shared, fast-moving research codebase.\n\nPREFERRED / BONUS EXPERIENCE\n• PhD in Machine Learning, Speech/Audio Processing, or Computer Science with publications at Interspeech, ICASSP, NeurIPS, ICML, or ICLR.\n• Direct hands-on experience training full-duplex speech-to-speech models or distilling autoregressive/flow audio models for sub-100ms streaming inference.",
      "updated_at": "2026-11-15T17:28:25.866Z",
      "created_at": "2026-11-15T16:53:33.574Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-audio-ml-research-engineer-speech-to-speech",
        "markdown": "https://jobs.talent247.co/jobs/senior-audio-ml-research-engineer-speech-to-speech.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-audio-ml-research-engineer-speech-to-speech/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-audio-ml-research-engineer-speech-to-speech/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "14712460-da7f-4aa6-9709-09758d9d521e",
      "slug": "machine-learning-engineer-multilingual-speech",
      "title": "Machine Learning Engineer, Multilingual Speech & Dubbing",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Edinburgh / Remote (UK)",
      "workplace_type": "Remote",
      "employment_type": "Full-time",
      "is_remote": true,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 90000,
        "max": 125000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£90k – £125k",
        "subtitle": "annual package"
      },
      "summary": "Machine Learning Engineer, Multilingual Speech & Dubbing (Machine Learning) — Edinburgh / Remote (UK) · Remote · £90k – £125k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar generative audio unicorn with a major UK engineering and research hub in London alongside distributed speech talent across the UK (including Edinburgh). Their voice synthesis, Automated Dubbing, speech-to-text, and conversational audio models power millions of creators, media studios, publishers, and global enterprises translating and voicing content across 70+ languages.\n\nThey have retained Talent247 to hire a Machine Learning Engineer, Multilingual Speech & Dubbing (Remote UK or Edinburgh hub). Joining their Automated Dubbing & Speech Understanding squad, you will own the models and massive-scale audio pipelines responsible for speaker diarization, overlapping-speech separation, cross-lingual prosody/emotion transfer, and frame-accurate lip/timing alignment across diverse global languages.\n\nKEY RESPONSIBILITIES\n• Train and iterate deep learning models for multi-speaker diarization, vocal-stem/background source separation, and noisy-environment multilingual ASR across broadcast, film, and conversational audio.\n• Develop duration-aware neural machine translation and cross-lingual prosody-transfer pipelines so dubbed speech preserves each original speaker's timbre, emotional delivery, and exact syllable/pause timing.\n• Curate, clean, and align hundreds of thousands of hours of multilingual audio-video datasets across high- and low-resource languages using automated quality filters and forced-alignment models.\n• Design perceptual and objective evaluation benchmarks (DER, WER, speaker-similarity cosine metrics, isochrony/lip-sync drift, and human MUSHRA listening panels) driving every model release.\n• Optimize PyTorch audio preprocessing and multi-model inference graphs to slash per-minute media dubbing latency and GPU compute cost.\n• Partner closely with product engineers and foundational TTS researchers to ship studio-grade dubbing controls to millions of users.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack uses Python, PyTorch, torchaudio, custom neural audio codecs/flow-matching models, Ray/Spark petabyte audio pipelines, C++/Rust media services, and H100 GPU clusters. Offered fully remote anywhere in the UK (with optional hub access in Edinburgh or London), compensation includes £90,000–£125,000 GBP base salary plus high-growth unicorn equity, home-office hardware budget, private healthcare, and generous PTO.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 3–6+ years of hands-on machine learning engineering experience in speech processing, audio ML, ASR, speaker diarization, source separation, or speech translation.\n• High proficiency in Python and PyTorch, including end-to-end training of sequence/conformer/transformer or diffusion audio architectures on large-scale speech datasets.\n• Solid signal-processing and acoustic fundamentals (mel-spectrograms, neural audio codecs, forced alignment, VAD, prosody/pitch modeling).\n• Practical experience building scalable audio dataset curation, filtering, and distributed preprocessing pipelines over tens of thousands of hours of speech.\n• Product-minded autonomy—comfortable owning features from dataset assembly and model training all the way to production API deployment.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience with automated video/audio dubbing, cross-lingual voice cloning, or multimodal audio-visual lip synchronization.\n• Multilingual proficiency or linguistic familiarity with non-English phonetic/tonal systems, plus publications at Interspeech or ICASSP.",
      "updated_at": "2026-11-15T17:28:03.429Z",
      "created_at": "2026-11-15T16:53:40.659Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech",
        "markdown": "https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7d65e4c6-8d14-4987-b04b-1e6f76de8287",
      "slug": "senior-spoken-dialogue-small-language-model",
      "title": "Senior Spoken Dialogue & Small Language Model Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 135000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£95k – £135k",
        "subtitle": "annual package"
      },
      "summary": "Senior Spoken Dialogue & Small Language Model Engineer (Machine Learning) — London, UK · Hybrid · £95k – £135k. THE CLIENT & MANDATE Our client, spun out…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client, spun out of Cambridge's renowned spoken-language systems research lab and headquartered in London, is Europe's category leader in enterprise conversational voice agents. Serving hundreds of global brands across banking, insurance, healthcare, and hospitality, their platform must resolve complex multi-turn spoken transactions—authentication, alphanumeric account codes, interruptions, regional accents, and noisy mobile lines—within strict sub-500ms conversational response budgets.\n\nThey have partnered with Talent247 to appoint a Senior Spoken Dialogue & Small Language Model Engineer in London. Sitting inside their Core Dialogue & Modelling R&D group, you will post-train specialized small language models (SLMs) and joint speech-text encoders engineered specifically for spoken turn-taking, zero-hallucination tool execution, and robust slot extraction over imperfect ASR hypotheses.\n\nKEY RESPONSIBILITIES\n• Post-train and distil compact 1B–14B spoken-dialogue language models using SFT, DPO, and online RL from simulated caller rollouts to achieve frontier-level reasoning inside sub-150ms token-generation budgets.\n• Design ASR-error-resilient architectures—consuming N-best hypotheses, phoneme/acoustic embeddings, and prosodic turn-end cues—to eliminate premature interruptions and misheard alphanumeric entities (postcodes, policy IDs, dates).\n• Architect constrained tool-calling and schema-guided decoding layers guaranteeing that live voice agents never fabricate enterprise policies, pricing, or account actions.\n• Build multi-turn conversational simulation environments where adversarial synthetic user agents test complex clarification, repair, and escalation strategies across dozens of languages and dialects.\n• Collaborate closely with GPU inference engineers to quantize and serve multi-LoRA tenant adapters with high prefix-cache efficiency on shared vLLM/TensorRT-LLM clusters.\n• Track and publish applied advances in spoken dialogue systems, turn-taking modeling, and task-oriented LLM alignment.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir R&D stack spans Python, PyTorch, FSDP/DeepSpeed, TRL/verl, vLLM/TensorRT-LLM, streaming speech encoders, and multi-GPU AWS/dedicated clusters. Working hybrid in West London, our client offers £95,000–£135,000 GBP base salary plus generous Series D stock options, private medical insurance, pension match, and UK Skilled Worker / Global Talent visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in NLP, conversational AI, speech-language modeling, or LLM post-training engineering.\n• Deep hands-on proficiency in PyTorch, transformer fine-tuning/distillation (SFT, LoRA, DPO, RLHF/GRPO), and low-latency small-language-model (SLM) optimization.\n• Strong understanding of spoken-language phenomena—ASR noise robustness, endpointing/turn-taking, disfluencies, coreference, and multi-turn dialogue state tracking.\n• Proven track record building rigorous offline and simulated multi-turn evaluation suites that correlate reliably with live production call containment and CSAT.\n• Clean software engineering habits in a shared Python/C++ research-to-production codebase.\n\nPREFERRED / BONUS EXPERIENCE\n• PhD or MSc in Spoken Dialogue Systems, Computational Linguistics, Speech Recognition, or Machine Learning (Interspeech, ICASSP, ACL, EMNLP publications a plus).\n• Experience with joint audio-text multimodal LLMs or speculative/grammar-constrained decoding in vLLM/TensorRT-LLM.",
      "updated_at": "2026-11-15T17:27:54.090Z",
      "created_at": "2026-11-15T16:53:44.221Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-spoken-dialogue-small-language-model",
        "markdown": "https://jobs.talent247.co/jobs/senior-spoken-dialogue-small-language-model.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-spoken-dialogue-small-language-model/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-spoken-dialogue-small-language-model/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "1560fdf5-e8e3-49d9-a90c-47e4c81c3295",
      "slug": "senior-forward-deployed-voice-ai-engineer",
      "title": "Senior Forward-Deployed Voice AI Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 85000,
        "max": 115000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£85k – £115k",
        "subtitle": "annual package"
      },
      "summary": "Senior Forward-Deployed Voice AI Engineer (Applied AI) — London, UK · Hybrid · £85k – £115k. THE CLIENT & MANDATE Our client is a Series D conversational…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a Series D conversational voice AI scale-up headquartered in West London, founded by spoken-dialogue pioneers from Cambridge's Machine Intelligence Laboratory. Their enterprise voice agents handle tens of millions of live, inbound customer phone calls annually for major retail banks, airlines, hotel groups, utilities, and healthcare networks—operating in high-stakes environments where latency spikes or hallucinated policy answers immediately impact thousands of callers.\n\nThey have retained Talent247 to hire a Senior Forward-Deployed Voice AI Engineer in London. Combining 70% hands-on Python/TypeScript AI systems engineering with 30% direct technical leadership alongside enterprise architecture teams, you will take complex multi-turn voice agent deployments from initial technical discovery through sub-second telephony integration, automated simulation testing, and millions of live production calls.\n\nKEY RESPONSIBILITIES\n• Own end-to-end technical architecture and implementation of bespoke enterprise voice agents, designing multi-turn dialogue state machines, LLM tool-calling flows, and deterministic policy guardrails.\n• Build resilient, low-latency REST, GraphQL, and webhook integrations connecting live spoken conversations with enterprise CRM, core banking, booking, identity-verification, and contact-center (SIP/CCaaS) platforms.\n• Design synthetic caller-persona simulation suites and automated LLM rubric evaluators that stress-test accents, barge-ins, noisy backgrounds, and edge-case business rules prior to go-live.\n• Analyze live call transcripts, latency traces, and containment telemetry to systematically drive task-completion rates higher while eliminating conversational friction.\n• Partner directly with core ML dialogue, ASR, and TTS researchers in London to surface real-world acoustic and linguistic edge cases and validate new model releases in production.\n• Extract repeatable patterns from customer deployments into reusable SDK modules, agent templates, and internal developer tooling.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack combines Python, TypeScript, custom small-language-model (SLM) & frontier LLM dialogue engines, streaming ASR/TTS, SIP/WebRTC telephony, Kubernetes on AWS, and automated simulation harnesses. Based hybrid in London, our client offers £85,000–£115,000 GBP base salary plus Series D equity options, private healthcare, 25+ days holiday, and UK visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of software, applied AI, or forward-deployed engineering experience building and shipping production backend and LLM/conversational systems.\n• Strong coding proficiency in Python and TypeScript, with solid command of asynchronous APIs, distributed tracing, automated testing, and clean system design.\n• Hands-on experience orchestrating LLMs in production—including structured tool calling, prompt/context engineering, deterministic guardrails, and automated evaluation pipelines.\n• Proven ability to lead technical architecture discussions directly with enterprise engineering, security, and operations stakeholders under real launch timelines.\n• Sharp analytical instincts for debugging live conversational logs, latency breakdowns, and API edge cases.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience with voice/telephony stacks (WebRTC, SIP, Twilio/Genesys/Amazon Connect, streaming ASR/TTS, VAD/turn-taking).\n• Experience working in regulated enterprise verticals such as financial services, healthcare, telecoms, or travel.",
      "updated_at": "2026-11-15T17:27:41.960Z",
      "created_at": "2026-11-15T16:53:47.633Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-forward-deployed-voice-ai-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-forward-deployed-voice-ai-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-forward-deployed-voice-ai-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-forward-deployed-voice-ai-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "60e4d961-c762-4d20-a1ac-29446a0d12b7",
      "slug": "principal-sovereign-gpu-cloud-infrastructure",
      "title": "Principal Sovereign GPU Cloud Infrastructure Architect",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "London, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 130000,
        "max": 175000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£130k – £175k",
        "subtitle": "annual package"
      },
      "summary": "Principal Sovereign GPU Cloud Infrastructure Architect (AI Infrastructure) — London, UK · Hybrid · £130k – £175k. THE CLIENT & MANDATE Our client is one…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of Europe's largest and fastest-growing AI infrastructure unicorns ($1B+ capital raised), headquartered in London and building vertically integrated, liquid-cooled sovereign GPU data centers from the ground up. They provide frontier model laboratories, national sovereign AI programmes, and global enterprises with dedicated multi-thousand-GPU training clusters and managed Kubernetes AI clouds free from hyperscaler virtualization tax.\n\nThey have engaged Talent247 to appoint a Principal Sovereign GPU Cloud Infrastructure Architect in London. Reporting into senior infrastructure leadership as a high-leverage technical authority, you will define the end-to-end bare-metal-to-Kubernetes architecture governing how tens of thousands of next-generation accelerators (HGX H200, GB200 NVL72) are provisioned, networked, isolated for strict multi-tenancy, and kept at >98% goodput.\n\nKEY RESPONSIBILITIES\n• Define and evolve the reference compute, storage, and fabric architecture for multi-site bare-metal Kubernetes and Slurm GPU superclusters scaling from thousands to tens of thousands of accelerators.\n• Architect zero-touch bare-metal lifecycle automation—spanning PXE/network boot, BMC/Redfish firmware orchestration, immutable Linux/driver imaging, automated burn-in validation, and silent-data-corruption quarantine.\n• Design rail-optimized 400G/800G InfiniBand and RoCEv2 lossless fabrics alongside L2/L3 cryptographic tenant isolation, SR-IOV/GPUDirect RDMA passthrough, and sovereign enclave compliance boundaries.\n• Partner with distributed storage leads to engineer multi-petabyte parallel filesystem and caching tiers (Lustre, WEKA, VAST, Ceph) delivering 20–50 GB/s per node for checkpointing and weight loading.\n• Lead cross-team architectural reviews, hardware vendor co-design sessions, and root-cause analyses on complex hardware-kernel-network interactions.\n• Mentor staff and senior engineers, establishing reusable operators and operational standards across global data center expansions.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack spans bare-metal Linux, Kubernetes custom controllers (Go/Rust), Slurm, NVIDIA GB200 NVL72 / HGX H200, InfiniBand/SONiC/EVPN, Redfish/BMC automation, WEKA/VAST parallel storage, and Terraform. Based hybrid in central London, our client offers £130,000–£175,000 GBP base salary plus substantial unicorn equity, executive benefits, and UK visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 8+ years of infrastructure, HPC, neocloud, or hyperscaler systems architecture experience, including proven technical leadership over large-scale bare-metal GPU or supercomputing estates.\n• Deep hands-on mastery of NVIDIA multi-node GPU architectures (HGX H100/H200, NVLink/NVSwitch, GB200 NVL72, MIG, DCGM) and GPUDirect RDMA networking (InfiniBand or RoCEv2).\n• Strong systems engineering track record building Kubernetes bare-metal operators/controllers (Cluster API, Metal3, Kueue, Volcano) and/or Slurm orchestration in Go, Rust, or Python.\n• Thorough command of Linux kernel/driver internals, BMC/Redfish hardware lifecycle tooling, and multi-tenant cloud security isolation.\n• Exceptional architectural communication skills across data center hardware, networking, security, and ML research stakeholders.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience commissioning liquid-cooled GB200 rack-scale clusters or designing sovereign/government-accredited cloud enclaves.\n• Hands-on familiarity with NCCL collective profiling and parallel AI storage fabrics (WEKA, VAST, Lustre).",
      "updated_at": "2026-11-15T17:27:32.396Z",
      "created_at": "2026-11-15T16:53:50.684Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure",
        "markdown": "https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "4d019898-5d40-4fe6-a550-709f67d7fd03",
      "slug": "senior-distributed-llm-inference-platform-engineer",
      "title": "Senior Distributed LLM Inference Platform Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Bristol, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 130000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£95k – £130k",
        "subtitle": "annual package"
      },
      "summary": "Senior Distributed LLM Inference Platform Engineer (AI Infrastructure) — Bristol, UK · Hybrid · £95k – £130k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a billion-dollar sovereign AI hyperscaler headquartered in the UK, deploying vertically integrated, renewable-powered GPU superclusters across Europe and North America. On top of their bare-metal accelerator fleet, they operate a high-throughput serverless inference and fine-tuning platform serving frontier open-weight language, reasoning, and multimodal models to AI-native startups and regulated European enterprises.\n\nThey have retained Talent247 to hire a Senior Distributed LLM Inference Platform Engineer at their Bristol systems engineering hub. Tapping into Bristol's deep parallel-computing and silicon-software heritage, you will architect the multi-node disaggregated serving layer, KV-cache tiering fabric, and intelligent autoscaling router powering trillions of served tokens across their GPU cloud.\n\nKEY RESPONSIBILITIES\n• Architect, deploy, and optimize multi-node distributed LLM serving clusters (vLLM, SGLang, TensorRT-LLM, NVIDIA Dynamo) supporting 100B–600B+ parameter MoE and long-context reasoning models.\n• Implement prefill/decode disaggregation, RDMA/NVLink KV-cache transfer, distributed prefix-cache routing, and speculative decoding across multi-rack H100/H200/GB200 pods.\n• Resolve systemic performance and reliability bottlenecks from GPU kernel stalls and NCCL collective hangs up to Envoy/Rust request scheduling and multi-tenant noisy-neighbor isolation.\n• Build reproducible, automated benchmarking harnesses quantifying TTFT, TPOT, goodput under strict SLOs, and tokens-per-megawatt/dollar across newly released open-weight models and hardware generations.\n• Evaluate emerging serving engines, quantization kernels (FP8/NVFP4), and RL rollout runtimes, contributing fixes upstream where impactful.\n• Collaborate closely with bare-metal Kubernetes infrastructure and product teams to turn raw cluster capacity into effortless developer APIs.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir serving stack combines Python, C++, Rust, Go, PyTorch, vLLM, SGLang, TensorRT-LLM, InfiniBand/RoCEv2 RDMA, Kubernetes, and OpenTelemetry. Working hybrid in Bristol, our client offers £95,000–£130,000 GBP base salary plus meaningful hyperscaler equity, private medical insurance, generous holiday allowance, and flexible human-first working practices.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems, ML infrastructure, or distributed backend engineering experience operating performance-critical production platforms.\n• Direct hands-on experience deploying, profiling, and scaling modern LLM inference engines (vLLM, SGLang, TensorRT-LLM, or Triton) across multi-GPU Kubernetes clusters.\n• Strong programming proficiency in Python and at least one systems language (C++, Rust, or Go).\n• Deep understanding of LLM serving internals: continuous batching, PagedAttention, KV-cache memory hierarchy, FP8/INT4 quantization, and tensor/expert parallelism.\n• Data-driven engineering discipline using Nsight, PyTorch Profiler, Prometheus, and distributed tracing to back architectural tradeoffs with hard numbers.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with disaggregated inference routing (NVIDIA Dynamo, Mooncake, LMCache) or GPUDirect RDMA / Nixl KV-cache transport.\n• Upstream contributions to open-source LLM inference engines or kernel libraries.",
      "updated_at": "2026-11-15T17:27:21.008Z",
      "created_at": "2026-11-15T16:53:54.328Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-distributed-llm-inference-platform-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-distributed-llm-inference-platform-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-distributed-llm-inference-platform-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-distributed-llm-inference-platform-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "fbe5ed61-aabd-4fa4-9cf2-f4605baf8615",
      "slug": "senior-research-scientist-generative-materials-ai",
      "title": "Senior Research Scientist, Generative Materials AI",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Cambridge, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 115000,
        "max": 160000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£115k – £160k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Scientist, Generative Materials AI (AI Research) — Cambridge, UK · Hybrid · £115k – £160k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a $2B+-valued Cambridge, UK deep-tech unicorn founded by world-renowned pioneers in geometric deep learning and computational chemistry. Backed by premier AI chipmakers, industrial partners, and tier-one venture funds, they build generative foundation models and universal neural interatomic potentials capable of proposing and evaluating entirely new crystalline, porous, and molecular materials orders of magnitude faster than traditional trial-and-error synthesis.\n\nThey have partnered with Talent247 to appoint a Senior Research Scientist, Generative Materials AI in Cambridge. Working alongside leading physicists, chemists, and ML theorists, you will invent next-generation equivariant diffusion/flow-matching architectures and property-guided generation algorithms that design synthesizable materials meeting exact real-world industrial specifications.\n\nKEY RESPONSIBILITIES\n• Conceive, design, and pre-train generative models (3D crystal diffusion, Riemannian flow matching, SE(3)/E(3)-equivariant transformers, autoregressive space-group models) over millions of inorganic, metal-organic, and molecular structures.\n• Advance universal neural force fields / machine-learned interatomic potentials (MLIPs) predicting energy, forces, stress tensors, phonons, and free-energy stability at quantum-chemical accuracy and millisecond speed.\n• Develop reinforcement-learning, classifier-free guidance, and synthesizability-constrained generation methods steering models toward manufacturable, low-cost, non-toxic elemental compositions.\n• Collaborate closely with quantum-simulation engineers and experimental synthesis partners to run closed-loop prospective discovery campaigns in carbon capture, solid-state electrolytes, and semiconductor precursors.\n• Scale clean, well-tested JAX or PyTorch training and evaluation suites across multi-node H100 clusters.\n• Publish high-impact research at NeurIPS, ICML, ICLR, and Nature-family journals while mentoring junior researchers.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers work in Python, JAX, and PyTorch with dedicated multi-node GPU clusters, automated DFT active-learning loops, and world-class academic/industrial laboratory partnerships. Based hybrid in historic Cambridge, UK, our client offers £115,000–£160,000 GBP base salary plus substantial unicorn equity, comprehensive UK health and pension benefits, and Global Talent / Skilled Worker visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computational Physics, Quantum Chemistry, Materials Science, or Applied Mathematics with a strong publication record.\n• First-author papers at top ML or scientific venues (NeurIPS, ICML, ICLR, Nature Computational Science, npj Computational Materials, JACS, or Physical Review Letters).\n• Deep hands-on expertise in geometric deep learning, equivariant graph neural networks, 3D diffusion/flow matching, or neural interatomic potentials in PyTorch or JAX.\n• Solid grounding in solid-state physics, crystallography, statistical mechanics, or quantum chemistry (DFT).\n• Demonstrated ability to write scalable distributed training code and validate generative models on physically meaningful downstream tasks.\n\nPREFERRED / BONUS EXPERIENCE\n• Track record contributing to widely adopted open-source atomistic ML frameworks or benchmarks.\n• Experience modeling synthesizability, phase diagrams, or transport/adsorption properties in porous or crystalline materials.",
      "updated_at": "2026-11-15T17:27:10.038Z",
      "created_at": "2026-11-15T16:53:56.990Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-scientist-generative-materials-ai",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-scientist-generative-materials-ai.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-generative-materials-ai/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-generative-materials-ai/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "16c19a1c-2533-4c32-b0ba-b6093f868ce2",
      "slug": "senior-mlops-scientific-simulation-pipeline",
      "title": "Senior MLOps & Scientific Simulation Pipeline Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Cambridge, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 90000,
        "max": 125000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£90k – £125k",
        "subtitle": "annual package"
      },
      "summary": "Senior MLOps & Scientific Simulation Pipeline Engineer (MLOps & Platform) — Cambridge, UK · Hybrid · £90k – £125k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a unicorn AI-for-materials discovery company headquartered in Cambridge, UK, backed by global semiconductor, climate, and deep-tech investors to design novel materials for carbon capture, next-generation batteries, and sustainable semiconductor manufacturing. Their platform operates like a \"search engine for the periodic table\"—combining generative crystal foundation models with learned neural force fields and massive quantum-mechanical Density Functional Theory (DFT) simulation campaigns.\n\nThey have retained Talent247 to hire a Senior MLOps & Scientific Simulation Pipeline Engineer in Cambridge. Joining their Compute & Simulation Infrastructure team, you will build the unified distributed orchestration engine coupling multi-node GPU generative screening with hundreds of thousands of concurrent HPC quantum-chemistry evaluations in a continuous active-learning loop.\n\nKEY RESPONSIBILITIES\n• Architect, scale, and operate resilient hybrid Kubernetes + Slurm workflows orchestrating petabyte-scale generative crystal generation, neural force-field relaxation, and high-throughput DFT quantum-simulation verification.\n• Design fault-tolerant active-learning state machines (Ray, Dagster/Prefect, Temporal) that automatically select uncertain candidate structures, dispatch quantum calculations, parse outputs, and trigger incremental foundation-model retraining.\n• Build high-performance scientific data lakes (Parquet, Lance, Zarr, PostgreSQL) indexing hundreds of millions of 3D atomic trajectories, electronic-structure tensors, and experimental synthesis results with full lineage.\n• Optimize GPU and high-core-count CPU cluster utilization, spot-instance checkpoint recovery, container packaging, and cost-per-evaluated-material metrics across cloud and supercomputing allocations.\n• Partner closely with ML research scientists and computational physicists to profile JAX/PyTorch training bottlenecks and package fast inference APIs for interactive materials exploration.\n• Establish CI/CD, reproducibility gates, and observability across the entire scientific ML lifecycle.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack features Python, PyTorch/JAX, Ray, Dagster/Prefect, Kubernetes, Slurm, AWS/GCP + sovereign HPC clusters, Apache Arrow/Parquet, and Terraform. Working hybrid in central Cambridge, our client offers £90,000–£125,000 GBP base salary plus unicorn equity options, private medical insurance, generous pension match, and UK visa/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in MLOps, ML platform engineering, scientific computing infrastructure, or distributed data/HPC engineering.\n• Strong Python software engineering skills combined with production mastery of modern workflow and distributed-compute orchestrators (Ray, Prefect, Dagster, Flyte, Airflow, or Temporal).\n• Hands-on operational command of Kubernetes and/or Slurm GPU/CPU clusters across AWS, GCP, or bare-metal HPC environments.\n• Experience designing high-throughput data pipelines, columnar/tensor storage schemas, model registries, and observability dashboards for PyTorch or JAX research teams.\n• Pragmatic builder mindset excited to turn messy scientific compute workflows into rock-solid, self-healing production systems.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior exposure to computational chemistry, materials science, molecular dynamics, or bioinformatics workflows (ASE, Pymatgen, VASP, Quantum ESPRESSO, OpenMM, RDKit).\n• Experience with Rust or C++ extensions for fast scientific data parsing and serialization.",
      "updated_at": "2026-11-15T17:26:57.855Z",
      "created_at": "2026-11-15T16:54:01.583Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-mlops-scientific-simulation-pipeline",
        "markdown": "https://jobs.talent247.co/jobs/senior-mlops-scientific-simulation-pipeline.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-mlops-scientific-simulation-pipeline/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-mlops-scientific-simulation-pipeline/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "33b9bebe-e4aa-4023-8001-ff7bf0ba2604",
      "slug": "senior-neural-simulation-synthetic-data-engineer",
      "title": "Senior Neural Simulation & Synthetic Data Engineer (Contract)",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "Oxford, UK (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Contract",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 130000,
        "currency": "GBP",
        "frequency": "yearly",
        "formatted": "£95k – £130k",
        "subtitle": "annual package"
      },
      "summary": "Senior Neural Simulation & Synthetic Data Engineer (Contract) (Robotics & Embodied AI) — Oxford, UK · Hybrid · £95k – £130k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an acclaimed autonomous-mobility software company spun out of Oxford's world-leading robotics institute, deploying safety-certified self-driving and industrial fleet autonomy software across commercial logistics hubs, airports, ports, and off-highway corridors globally. Because physical safety certification cannot rely solely on public-road mileage, their autonomy stack leans heavily on photorealistic neural scene reconstruction and physics-accurate synthetic sensor simulation.\n\nThey have retained Talent247 to engage a Senior Neural Simulation & Synthetic Data Engineer in Oxford on an initial 6-month contract (with strong likelihood of extension). Joining their Simulation & Synthetic Data group, you will scale 3D Gaussian Splatting / NeRF digital-twin pipelines and generative weather/hazard synthesis used to validate perception and planning stacks across millions of virtual miles.\n\nKEY RESPONSIBILITIES\n• Build and scale automated 3D scene-reconstruction pipelines (3D Gaussian Splatting, NeRFs, neural Lidar/radar fields) turning raw multi-camera and LiDAR field logs into editable, interactive simulation environments.\n• Develop generative augmentation workflows capable of inserting dynamic actors, rare industrial hazards, occlusion edge cases, and physically plausible rain, fog, snow, and glare into reconstructed scenes.\n• Optimize CUDA and C++/Python rendering and sensor-raycasting throughput so thousands of closed-loop regression scenarios execute rapidly in CI on GPU clusters.\n• Quantify synthetic-to-real perception fidelity by measuring downstream 3D object detection, occupancy-grid, and trajectory-planner correlation against physical test-track runs.\n• Partner closely with 3D perception ML engineers and safety-case leads to target long-tail failure modes systematically via automated active scenario mining.\n• Deliver clean, containerized, well-documented simulation services integrated into the core autonomy release pipeline.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack combines Python, C++17/20, PyTorch, CUDA, 3DGS/NeRF libraries, Unreal/Isaac Sim sensor interfaces, Docker, and Kubernetes GPU clusters. Operating hybrid in Oxford (2–3 days onsite), this contract mandate pays an annualized equivalent of £95,000–£130,000 GBP (structured as competitive contract/day-rate terms).",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of hands-on engineering or applied research experience in 3D computer vision, neural rendering, computer graphics, or autonomous-vehicle sensor simulation.\n• Strong practical expertise with 3D Gaussian Splatting (3DGS), Neural Radiance Fields (NeRFs), multi-view geometry/SLAM, or differentiable rendering in PyTorch and C++/CUDA.\n• Experience simulating or processing calibrated multi-camera, LiDAR, or radar sensor data for robotics or autonomous vehicles.\n• Solid software engineering skills across Python, C++, Docker, Linux, and GPU performance profiling.\n• Track record delivering production-ready simulation or synthetic-data tooling within cross-functional autonomy teams.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with NVIDIA Omniverse / Isaac Sim, Unreal Engine sensor plugins, or video/diffusion world models for driving scene generation.\n• Familiarity with safety-critical AV validation metrics and scenario description standards (OpenSCENARIO / OpenDRIVE).",
      "updated_at": "2026-11-15T17:26:38.596Z",
      "created_at": "2026-11-15T16:54:02.045Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-neural-simulation-synthetic-data-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-neural-simulation-synthetic-data-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-neural-simulation-synthetic-data-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-neural-simulation-synthetic-data-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3a8950d9-5177-41f3-9762-a45e2d2e3a1a",
      "slug": "research-scientist-frontier-pre-training-mixture",
      "title": "Research Scientist, Frontier Pre-Training & Mixture-of-Experts",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Paris, France (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 130000,
        "max": 185000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€130k – €185k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Frontier Pre-Training & Mixture-of-Experts (AI Research) — Paris, France · On-site · €130k – €185k. THE CLIENT & MANDATE Our client…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Europe's flagship frontier foundation model laboratory, headquartered in Paris and competing head-to-head at the global state of the art across multilingual reasoning, code generation, mathematics, and native multimodal understanding. With dedicated multi-thousand-GPU superclusters and a flat, researcher-led culture free of big-tech bureaucracy, small squads take bold architectural ideas from whiteboard scaling ablations directly to flagship global model launches.\n\nThey have partnered with Talent247 to recruit a Research Scientist, Frontier Pre-Training & Mixture-of-Experts onsite in Paris. Sitting inside their Core Pre-Training squad, you will co-own architectural design, optimizer dynamics, and multi-trillion-token data curricula for their next generation of sparse Mixture-of-Experts and native multimodal foundation models.\n\nKEY RESPONSIBILITIES\n• Design and validate novel sparse Mixture-of-Experts (MoE), hybrid state-space/attention, and native vision-audio-text transformer architectures from 1B-scale scaling-law sweeps up to flagship frontier training runs.\n• Investigate router load-balancing, expert specialization, auxiliary-loss-free routing, low-precision (FP8) training stability, and learning-rate/batch-size schedules across multi-thousand-GPU jobs.\n• Lead large-scale pre-training data science—optimizing multilingual, mathematical, scientific, and repository-level code mixtures, synthetic data upsampling, and mid-training long-context annealing (128k–1M+ tokens).\n• Write clean, high-efficiency distributed PyTorch code (3D/4D parallelism across tensor, pipeline, context, and expert dimensions) without relying on brittle black-box wrappers.\n• Monitor live frontier runs closely, diagnosing loss spikes, gradient norm anomalies, and downstream capability emergence in real time alongside infrastructure and post-training peers.\n• Contribute to landmark model releases and technical reports shaping the European and global AI landscape.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir research stack is built on custom high-performance PyTorch/CUDA/Triton distributed pre-training frameworks running on dedicated H100/H200/GB200 clusters. Based onsite at their central Paris headquarters, our client offers €130,000–€185,000 EUR base salary plus substantial frontier-lab BSPCE equity, full relocation/Talent Passport visa sponsorship, and top-tier compute-per-researcher ratios.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computer Science, Physics, or Mathematics, or equivalent hands-on track record pre-training large foundation models at a frontier AI lab.\n• Deep mathematical and empirical mastery of modern LLM architectures (GQA/MLA, RoPE/YaRN, RMSNorm, SwiGLU, sparse MoE routing) and scaling laws.\n• Ability to implement custom transformer/MoE layers, fused losses, and distributed sharding logic from scratch in PyTorch with complete command of tensor layouts and GPU memory budgets.\n• Direct experience running, debugging, and evaluating large-scale pre-training or continual pre-training experiments across multi-node GPU clusters.\n• First-author publications at NeurIPS, ICML, ICLR, or co-authorship on widely recognized frontier foundation model releases.\n\nPREFERRED / BONUS EXPERIENCE\n• Hands-on experience with FP8 pre-training recipes, multimodal vision/audio token pre-training, or custom Triton/CUDA training kernels.\n• Experience designing pre-training data filtering, classifier-based quality scoring, and synthetic pre-training corpora at trillion-token scale.",
      "updated_at": "2026-11-15T17:26:27.196Z",
      "created_at": "2026-11-15T16:54:33.408Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-frontier-pre-training-mixture",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-frontier-pre-training-mixture.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-frontier-pre-training-mixture/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-frontier-pre-training-mixture/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "66fa2b2a-d26f-4deb-8cf6-533e91c62fbe",
      "slug": "senior-cuda-distributed-llm-inference-engineer",
      "title": "Senior CUDA & Distributed LLM Inference Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 110000,
        "max": 155000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€110k – €155k",
        "subtitle": "annual package"
      },
      "summary": "Senior CUDA & Distributed LLM Inference Engineer (AI Infrastructure) — Paris, France · Hybrid · €110k – €155k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Europe's premier frontier foundation model laboratory headquartered in Paris, renowned globally for pioneering high-efficiency dense and sparse Mixture-of-Experts (MoE) language, code, and vision models. Because their models run everywhere—from massive multi-tenant cloud APIs to sovereign air-gapped defence clusters and single-node enterprise appliances—squeezing every last drop of memory bandwidth and compute utilization out of modern GPUs is existential to their platform advantage.\n\nThey have retained Talent247 to appoint a Senior CUDA & Distributed LLM Inference Engineer in Paris. Joining their core Inference Runtime & Kernels group, you will co-design low-level GPU kernels and distributed serving architectures alongside model researchers so newly trained architectures launch on day zero at world-leading throughput and minimal time-to-first-token.\n\nKEY RESPONSIBILITIES\n• Author, profile, and maintain custom C++20, CUDA/PTX, CUTLASS, and Triton kernels for grouped-query/sliding-window attention, sparse MoE routing/grouped GEMMs, and fused RoPE/SwiGLU/RMSNorm operators on Hopper and Blackwell GPUs.\n• Implement ultra-low-overhead quantization formats (FP8, NVFP4, INT4 weight-only and W4A8) paired with calibration pipelines that preserve reasoning and coding benchmark scores.\n• Scale multi-node disaggregated prefill/decode serving, expert parallelism over NVLink and InfiniBand RDMA, continuous batching, and radix prefix KV-cache sharing.\n• Collaborate upstream and internally across vLLM, TensorRT-LLM, and custom Rust/C++ serving runtimes to ensure sovereign on-premise and cloud API parity.\n• Partner with pre-training architects before major training runs lock in, modeling KV-cache footprint, expert-routing balance, and hardware FLOP utilization to ensure hardware-friendly architectures.\n• Build deterministic kernel correctness suites and hardware-level roofline profilers catching regressions on every commit.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir runtime stack spans C++20, CUDA/CUTLASS/PTX, Triton, Rust, Python/PyTorch, vLLM/TensorRT-LLM, NCCL/NVSHMEM, and dedicated H100/H200/B200 clusters. Working hybrid in central Paris, our client offers €110,000–€155,000 EUR base salary plus premier frontier-lab BSPCE equity, comprehensive benefits, and full visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of low-level GPU performance engineering, HPC systems, or ML inference runtime development.\n• Expert proficiency in C++17/20 and CUDA (including PTX/SASS inspection, CUTLASS/CuTe, Tensor Core MMA instructions, CUDA Graphs, and Nsight Compute/Systems profiling) and/or OpenAI Triton.\n• Deep production understanding of modern LLM serving mechanics: PagedAttention, continuous batching, speculative decoding, FP8/INT4 quantization, and tensor/pipeline/expert parallelism.\n• Strong grasp of multi-GPU interconnect behavior across NVLink, NVSwitch, PCIe, and GPUDirect RDMA / NCCL collectives.\n• Pragmatic, benchmark-driven engineering approach focused on end-to-end p99 latency and tokens-per-watt/dollar.\n\nPREFERRED / BONUS EXPERIENCE\n• Core contributions to vLLM, SGLang, TensorRT-LLM, FlashAttention, llama.cpp, or PyTorch CUDA internals.\n• Experience bringing up inference kernels on next-generation Blackwell (B200/GB200) or AMD MI300X architectures.",
      "updated_at": "2026-11-15T17:26:16.901Z",
      "created_at": "2026-11-15T16:54:36.251Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-cuda-distributed-llm-inference-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-cuda-distributed-llm-inference-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-cuda-distributed-llm-inference-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-cuda-distributed-llm-inference-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3cfba3d0-8317-4955-8e2a-4c31abdda255",
      "slug": "applied-ai-evaluation-engineer-enterprise",
      "title": "Applied AI & Evaluation Engineer, Enterprise Solutions",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 2,
      "salary": {
        "min": 90000,
        "max": 130000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€90k – €130k",
        "subtitle": "annual package"
      },
      "summary": "Applied AI & Evaluation Engineer, Enterprise Solutions (Applied AI) — Paris, France · Hybrid · €90k – €130k. THE CLIENT & MANDATE Our client is Europe's…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Europe's flagship decacorn frontier artificial intelligence laboratory, headquartered in Paris and backed by global technology and sovereign investors. They provide full-stack AI capabilities—spanning frontier open-weight and commercial reasoning/multimodal models, fine-tuning APIs, enterprise assistants, and dedicated sovereign compute—partnering directly with CAC 40, DAX 40, and Fortune 500 leaders across banking, defence, aerospace, energy, and telecoms.\n\nThey have engaged Talent247 to hire two Applied AI & Evaluation Engineers, Enterprise Solutions in Paris. Sitting at the technical vanguard between core model research and strategic enterprise transformations, you will co-architect custom fine-tuned models, domain evaluation suites, and compound agentic systems that run reliably inside high-stakes customer environments.\n\nKEY RESPONSIBILITIES\n• Partner directly with principal engineers at strategic European and global enterprise clients to translate mission-critical workflows into rigorous evaluation benchmarks, custom post-training recipes, and low-latency agent architectures.\n• Design and execute domain-specific model customization—spanning continual pre-training, SFT, LoRA/DPO, and RL with domain rubrics—to adapt compact 3B–24B and frontier MoE models to proprietary financial, industrial, and multilingual corpora.\n• Build automated, statistically sound evaluation harnesses (deterministic verifiers, LLM-as-a-judge rubrics, RAG faithfulness metrics, and latency/cost Pareto curves) that govern production launch sign-off.\n• Architect reference multi-agent tool-calling, structured-extraction, and hybrid retrieval pipelines deployed across cloud APIs, private VPCs, and on-premise GPU clusters.\n• Channel recurring capability gaps discovered in real enterprise workloads directly back to Paris pre-training and post-training research squads as clean benchmark tasks.\n• Build reusable internal SDKs, fine-tuning cookbooks, and evaluation templates that multiply field engineering velocity across Europe.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack features Python, PyTorch, TypeScript, vLLM/TensorRT-LLM, distributed fine-tuning pipelines, vector search, and dedicated GPU clusters. Based hybrid in central Paris, our client offers €90,000–€130,000 EUR base salary plus high-value frontier-lab equity (BSPCE), comprehensive French health coverage, and international travel to flagship partner engineering hubs.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of hands-on experience in machine learning engineering, applied AI research, or forward-deployed ML architecture shipping LLM systems into production.\n• Strong software engineering skills in Python (and working proficiency in TypeScript, Go, or Rust) with clean, modular, test-backed code habits.\n• Practical hands-on experience fine-tuning or post-training transformer models in PyTorch (SFT, PEFT/LoRA, DPO, synthetic data curation) and building automated LLM evaluation suites.\n• Deep architectural grasp of production RAG, structured JSON/grammar-constrained decoding, agentic tool calling, and latency-vs-accuracy routing tradeoffs.\n• Fluent professional English for executive and engineering collaboration (fluency in French, German, or another European language is a strong asset).\n\nPREFERRED / BONUS EXPERIENCE\n• Experience deploying quantized open-weight LLMs into regulated on-premise, VPC, or air-gapped enterprise environments.\n• MSc or PhD in Computer Science, Machine Learning, or Applied Mathematics from a leading technical university.",
      "updated_at": "2026-11-15T17:26:05.914Z",
      "created_at": "2026-11-15T16:54:40.181Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/applied-ai-evaluation-engineer-enterprise",
        "markdown": "https://jobs.talent247.co/jobs/applied-ai-evaluation-engineer-enterprise.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/applied-ai-evaluation-engineer-enterprise/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/applied-ai-evaluation-engineer-enterprise/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "ef24bb6a-9d39-4f04-9d48-ec3df4c00239",
      "slug": "senior-open-source-post-training-alignment-ml",
      "title": "Senior Open-Source Post-Training & Alignment ML Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 135000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€95k – €135k",
        "subtitle": "annual package"
      },
      "summary": "Senior Open-Source Post-Training & Alignment ML Engineer (Machine Learning) — Paris, France · Hybrid · €95k – €135k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is the Paris-anchored steward of the open-source AI ecosystem whose model hub, dataset repositories, and post-training libraries form the daily foundation for hundreds of thousands of ML engineers and academic researchers worldwide. Whenever a new alignment paper, reasoning RL algorithm, or memory-saving kernel drops on arXiv, their open-source post-training team turns it into clean, reproducible, multi-GPU production code adopted by the entire industry within days.\n\nThey have retained Talent247 to recruit a Senior Open-Source Post-Training & Alignment ML Engineer in Paris. Joining their Open Post-Training & Alignment Libraries team, you will co-maintain flagship open-source alignment toolkits, run reference post-training recipes on dedicated H100 clusters, and democratize frontier reasoning and multimodal RL techniques for the global builder community.\n\nKEY RESPONSIBILITIES\n• Design, implement, and maintain modular, high-performance PyTorch implementations of state-of-the-art post-training algorithms—including SFT, DPO, ORPO, GRPO, online RLOO/PPO, and verifiable-reward reasoning loops—across text and vision-language models.\n• Integrate memory- and compute-efficiency breakthroughs (FSDP2, DeepSpeed ZeRO, LoRA/QLoRA, Liger/Triton fused cross-entropy kernels, sequence packing, and vLLM async rollout co-location) so users train larger models on fewer GPUs.\n• Execute end-to-end open post-training reproductions on multi-node H100 Slurm clusters, publishing open recipes, training logs, datasets, and aligned checkpoints that match or beat closed baselines.\n• Build clean integration bridges between post-training trainers, sandboxed code/math verifiers, and standardized evaluation harnesses.\n• Review community pull requests, shape clean Pythonic APIs used by millions of monthly downloads, and author clear technical documentation and research blog posts.\n• Collaborate directly with hardware vendors and open-model labs to ensure day-zero post-training support for newly released model architectures.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack centers on Python, PyTorch, FSDP/Accelerate/DeepSpeed, Triton fused kernels, vLLM, Slurm H100 clusters, and open GitHub development. Working hybrid in Paris, our client offers €95,000–€135,000 EUR base salary plus equity options, full French benefits, conference travel, and global visibility for your technical work.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of hands-on machine learning engineering or applied research experience with deep expertise in PyTorch internals and transformer training dynamics.\n• Direct experience implementing and debugging LLM post-training methods (SFT, DPO, RLHF, GRPO/PPO, distillation) across multi-GPU / multi-node setups.\n• Proven open-source library authorship or core maintainership—you care deeply about clean abstractions, backward compatibility, unit tests, and crisp documentation.\n• Strong command of distributed GPU memory optimization (mixed precision BF16/FP8, activation checkpointing, FSDP/DeepSpeed, PEFT/quantization).\n• Clear technical writing skills for authoring reproducible recipes and engaging constructively with open-source contributors.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience co-locating fast inference engines (vLLM/SGLang) inside online RL training loops or writing custom Triton kernels for post-training losses.\n• Public portfolio of widely used open-source ML repos, aligned model releases, or technical papers/reports.",
      "updated_at": "2026-11-15T17:25:52.586Z",
      "created_at": "2026-11-15T16:54:43.330Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-open-source-post-training-alignment-ml",
        "markdown": "https://jobs.talent247.co/jobs/senior-open-source-post-training-alignment-ml.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-open-source-post-training-alignment-ml/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-open-source-post-training-alignment-ml/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "ac309492-3e9b-4df2-b1ba-bd56ad87f07b",
      "slug": "senior-rust-storage-ml-infrastructure-engineer",
      "title": "Senior Rust Storage & ML Infrastructure Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 90000,
        "max": 130000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€90k – €130k",
        "subtitle": "annual package"
      },
      "summary": "Senior Rust Storage & ML Infrastructure Engineer (MLOps & Platform) — Paris, France · Hybrid · €90k – €130k. THE CLIENT & MANDATE Our client operates the…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client operates the central open-source machine learning platform of the global AI ecosystem from their European headquarters in Paris, hosting millions of open-weight model repositories, petabyte-scale training datasets, and interactive ML demo spaces for millions of AI builders and enterprises worldwide. At their scale, content-addressed storage efficiency and multi-gigabyte-per-second weight transfer speeds directly govern how fast the entire AI community trains and deploys models.\n\nThey have partnered with Talent247 to hire a Senior Rust Storage & ML Infrastructure Engineer in Paris. Joining their Core Hub & Storage Systems team, you will engineer the high-throughput, content-addressed chunking protocol, Rust data plane, and global edge caching infrastructure that serves tens of petabytes of safetensors weights and Parquet shards every week.\n\nKEY RESPONSIBILITIES\n• Design, build, and scale high-performance Rust storage and transfer services handling content-defined chunking, block-level deduplication, and multi-stream parallel uploads/downloads of 100GB+ model checkpoints and terabyte datasets.\n• Optimize zero-copy HTTP/2, gRPC, and presigned CDN data paths so distributed GPU clusters saturate 10–40 Gbps per node when pulling model weights or streaming Parquet training shards.\n• Architect metadata indexing, garbage collection, cryptographic integrity verification, and repository versioning semantics across billions of stored objects.\n• Partner with open-source client SDK maintainers (Python and Rust) to ensure seamless, resilient transfer performance across CLI tools, training libraries, and serverless inference containers.\n• Operate production infrastructure on Kubernetes and multi-cloud object stores with Prometheus/OpenTelemetry observability, automated chaos testing, and strict cost-per-terabyte efficiency.\n• Contribute openly to public protocols, technical blog posts, and open-source Rust crates used across the global ML community.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir infrastructure stack is built on Rust (Tokio, Axum, Hyper), Go, Python, S3-compatible multi-cloud storage, global CDNs, MongoDB/PostgreSQL, ClickHouse, and Kubernetes. Offered hybrid in Paris, compensation includes €90,000–€130,000 EUR base salary, meaningful equity options, generous French PTO, and an open, transparent engineering culture.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of backend and distributed storage or infrastructure engineering experience operating high-egress, latency-sensitive internet-scale services.\n• Strong production mastery of Rust (async Tokio, memory management, zero-copy I/O, FFI/Python bindings via PyO3) alongside working knowledge of Go or Python.\n• Deep familiarity with object storage internals (S3 APIs, multipart streaming, content-addressed storage, Merkle trees, deduplication algorithms) and CDN caching mechanics.\n• Solid operational command of Kubernetes, Linux network/TCP tuning, distributed observability, and high-availability rollouts.\n• Collaborative, low-ego communication style suited to open-source-first engineering.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with ML file formats (`safetensors`, Apache Parquet, Arrow, Lance) or Git/Git-LFS internals at scale.\n• Published open-source Rust crates or contributions to storage/networking projects.",
      "updated_at": "2026-11-15T17:25:38.547Z",
      "created_at": "2026-11-15T16:54:46.522Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-rust-storage-ml-infrastructure-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-rust-storage-ml-infrastructure-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-rust-storage-ml-infrastructure-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-rust-storage-ml-infrastructure-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7a40e7a8-eec0-4907-bbbc-caa82134793e",
      "slug": "senior-rl-sandboxed-code-execution-systems",
      "title": "Senior RL & Sandboxed Code Execution Systems Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 120000,
        "max": 170000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€120k – €170k",
        "subtitle": "annual package"
      },
      "summary": "Senior RL & Sandboxed Code Execution Systems Engineer (Machine Learning) — Paris, France · Hybrid · €120k – €170k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a $500M+-funded frontier AI company with its primary engineering and research hub in Paris, building custom foundation models engineered specifically for enterprise software development via Reinforcement Learning from Code Execution Feedback (RLCEF). In their training architecture, the bottleneck to smarter coding models isn't just GPU FLOPs—it's executing millions of heterogeneous, stateful multi-container repository builds, tests, and browser/API checks per minute safely and deterministically.\n\nThey have retained Talent247 to hire a Senior RL & Sandboxed Code Execution Systems Engineer in Paris. Joining their RL Infrastructure & Execution Environments team, you will own the planetary-scale microVM/container orchestration engine and asynchronous RL rollout pipeline connecting multi-thousand-GPU training clusters with real-time compiler and test-suite reward environments.\n\nKEY RESPONSIBILITIES\n• Architect, scale, and optimize high-throughput sandboxed execution infrastructure (Firecracker microVMs, gVisor, overlayfs/snapshot-restore) running millions of untrusted multi-language repository builds and test suites concurrently.\n• Build low-overhead asynchronous RL rollout engines (orchestrating vLLM/SGLang inference workers, environment steppers, and PyTorch FSDP/Megatron learners) maximizing GPU utilization during long-horizon agentic coding episodes.\n• Engineer hermetic build-caching, deterministic network/clock virtualization, and anti-tampering sandbox isolation so RL agents cannot game test harnesses or leak side effects.\n• Create automated repository ingestion pipelines that turn raw multi-service monorepos (C++, Rust, Java, TypeScript, Python, Go) into reproducible, sub-second-bootable RL environments.\n• Profile and eliminate tail-latency stragglers across distributed storage, container snapshotting, and RPC trajectory streaming.\n• Partner directly with RL research scientists to expose rich intermediate execution traces, coverage diffs, and profiler signals as dense reward inputs.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir infrastructure stack runs on Rust, Go, C++, Python, Linux kernel virtualization (KVM, eBPF, cgroups v2, namespaces), Kubernetes, vLLM, and dedicated GPU + high-core-count CPU superclusters. Working hybrid in central Paris, our client offers €120,000–€170,000 EUR base salary plus substantial equity, comprehensive family health coverage, and full relocation/visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of distributed systems, cloud runtime, or ML infrastructure engineering experience operating at massive concurrency.\n• Strong systems programming mastery in Rust, Go, or C++, alongside solid Python proficiency for ML/RL pipeline integration.\n• Deep hands-on expertise in Linux isolation and sandboxing primitives (Firecracker, KVM, gVisor, namespaces, cgroups, seccomp, eBPF, copy-on-write filesystems).\n• Experience building high-throughput job schedulers, CI/build execution engines (Bazel/Nix/containers), or distributed RL rollout frameworks (verl, Ray, OpenRLHF, or custom equivalents).\n• Ability to debug subtle kernel, filesystem, and distributed concurrency bugs that only surface at million-container scale.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience supporting frontier LLM post-training or SWE-agent reinforcement learning runs.\n• Familiarity with vLLM/SGLang continuous batching, KV-cache prefix sharing across tree rollouts, and RDMA/high-speed networking.",
      "updated_at": "2026-11-15T17:25:27.861Z",
      "created_at": "2026-11-15T16:54:50.050Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-rl-sandboxed-code-execution-systems",
        "markdown": "https://jobs.talent247.co/jobs/senior-rl-sandboxed-code-execution-systems.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-rl-sandboxed-code-execution-systems/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-rl-sandboxed-code-execution-systems/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "8e100787-891f-46e1-a0b7-89ae7bfeb028",
      "slug": "research-scientist-program-synthesis-code",
      "title": "Research Scientist, Program Synthesis & Code Reasoning",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 125000,
        "max": 175000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€125k – €175k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Program Synthesis & Code Reasoning (AI Research) — Paris, France · Hybrid · €125k – €175k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a $500M+-capitalized frontier AI laboratory operating its flagship R&D headquarters in Paris, purpose-built to solve autonomous software engineering from first principles. Rather than treating code generation merely as next-token text prediction over public repositories, they pre-train and post-train proprietary foundation models closed-loop against millions of live compiler, static-analyzer, and execution environments—deploying their models directly inside security-conscious global enterprises and defence-industrial organizations.\n\nThey have engaged Talent247 to appoint a Research Scientist, Program Synthesis & Code Reasoning in Paris. Working alongside world-class RL researchers, programming-language theorists, and supercomputing systems engineers, you will invent new training and test-time search paradigms that enable foundation models to reason soundly across million-line repositories, formal invariants, and complex multi-step architecture changes.\n\nKEY RESPONSIBILITIES\n• Pioneer reinforcement learning and test-time search algorithms (MCTS, verifier-guided beam/tree search, process reward models, RLVR) grounded in compiler diagnostics, type checkers, fuzzers, and unit/integration test execution.\n• Design synthetic repository evolution and bug-injection curricula that teach long-horizon repository navigation, cross-file refactoring, root-cause debugging, and security-vulnerability patching.\n• Explore neuro-symbolic bridges combining formal verification (Lean, Dafny, SMT solvers, abstract interpretation) with large-scale autoregressive and diffusion code models.\n• Run rigorous scaling-law ablations across data mixtures, context lengths (128k–1M+ tokens), and RL compute allocation on dedicated multi-thousand-GPU clusters.\n• Collaborate closely with distributed execution-sandbox engineers to ensure reward signals remain fast, deterministic, and immune to reward hacking or test tampering.\n• Translate breakthrough research directly into flagship enterprise model releases.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers work on a massive dedicated H100/H200 GPU cluster with custom PyTorch/JAX distributed training frameworks and a planet-scale sandboxed code-execution farm. Based hybrid in Paris (with relocation support across Europe and globally), our client offers €125,000–€175,000 EUR base salary plus top-decile frontier-lab equity, private healthcare, and full French Talent Passport visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Programming Languages, Formal Methods, or Computer Science, or equivalent frontier lab track record in LLM reasoning and code synthesis.\n• Publications at premier venues (NeurIPS, ICML, ICLR, POPL, PLDI, OOPSLA, CAV, or ICSE) or demonstrated leadership on frontier code/reasoning foundation models.\n• Strong hands-on proficiency in PyTorch or JAX for large-scale LLM pre-training, post-training (RLHF/GRPO/PPO), or neural search algorithms.\n• Deep familiarity with static program analysis, AST/CFG/SSA representations, type systems, compilers, or automated theorem proving / fuzzing.\n• Ability to design clean controlled experiments and collaborate tightly with systems engineers at thousand-GPU scale.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience mitigating reward hacking in verifiable code or math RL environments (e.g., SWE-bench, LiveCodeBench, repository-scale benchmarks).\n• Systems fluency in Rust or C++ alongside Python.",
      "updated_at": "2026-11-15T17:25:15.297Z",
      "created_at": "2026-11-15T16:54:53.516Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-program-synthesis-code",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-program-synthesis-code.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-program-synthesis-code/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-program-synthesis-code/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "56099269-1fe3-4e69-9430-153afcbe4a61",
      "slug": "senior-agentic-orchestration-engineer-typescript",
      "title": "Senior Agentic Orchestration Engineer (TypeScript / Rust)",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Paris, France (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 85000,
        "max": 120000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€85k – €120k",
        "subtitle": "annual package"
      },
      "summary": "Senior Agentic Orchestration Engineer (TypeScript / Rust) (Applied AI) — Paris, France · Hybrid · €85k – €120k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an elite, fast-scaling Paris AI company founded by repeat technical founders and former frontier-lab infrastructure leads. They build an extensible enterprise multi-agent operating system that lets engineering, support, finance, and operations teams compose custom AI agents securely wired into live company knowledge across Slack, Notion, Google Drive, GitHub, Linear, Zendesk, and cloud data warehouses.\n\nThey have retained Talent247 to hire a Senior Agentic Orchestration Engineer (TypeScript / Rust) at their central Paris headquarters. Joining a compact, ultra-senior product and infrastructure engineering team where every engineer owns end-to-end architecture from storage down to developer UX, you will scale the durable workflow engine, Model Context Protocol (MCP) tool broker, and high-throughput Rust ingestion layer powering hundreds of thousands of daily enterprise agent executions.\n\nKEY RESPONSIBILITIES\n• Architect and evolve the durable multi-agent execution harness (Temporal + TypeScript/Rust), supporting recursive sub-agent delegation, human-in-the-loop tool approvals, sandboxed code execution, and graceful provider failover across frontier LLMs.\n• Build and harden their enterprise MCP server and remote tool-execution gateway, enforcing fine-grained OAuth2 scopes, RBAC space permissions, and audit logging on every external write action.\n• Scale the high-concurrency Rust document ingestion, incremental sync, chunking, and Qdrant vector-indexing pipeline handling tens of millions of live SaaS documents and database rows.\n• Optimize streaming token latency, prompt-cache hit rates, context-window compaction, and structured-output parsing across heterogeneous model providers.\n• Ship crisp, responsive agent-builder and observability interfaces in TypeScript/React so technical and non-technical builders can inspect full execution traces and debug tool calls effortlessly.\n• Engage directly with forward-thinking enterprise engineering customers to turn real workflow bottlenecks into clean platform primitives.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack is TypeScript, Rust, React, Temporal, Qdrant, PostgreSQL, Redis, Kubernetes on GCP, and multi-provider frontier LLM + MCP APIs. Working hybrid in central Paris, our client offers €85,000–€120,000 EUR base salary plus generous early-growth BSPCE equity, French Mutuelle health coverage, meal vouchers, and visa sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of full-stack and backend systems engineering experience shipping high-reliability SaaS infrastructure or developer tooling at scale.\n• Deep production mastery of TypeScript (Node.js & React) and strong systems proficiency in Rust (or Go/C++ with eagerness to write production Rust daily).\n• Hands-on experience building stateful LLM agent loops, Model Context Protocol (MCP) servers/clients, durable workflow orchestration (Temporal or equivalent), or semantic RAG pipelines.\n• Solid distributed-systems fundamentals across PostgreSQL performance tuning, vector search (Qdrant/pgvector), OAuth2/webhook sync architectures, and rate-limit handling.\n• High autonomy, product taste, and clear written communication in an English-speaking engineering culture in Paris.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience building SaaS data connectors, sandboxed code-interpreter runtimes, or LLM evaluation/observability tooling.\n• Active open-source contributions in the TypeScript, Rust, or agentic tooling ecosystem.",
      "updated_at": "2026-11-15T17:25:03.333Z",
      "created_at": "2026-11-15T16:54:56.216Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-agentic-orchestration-engineer-typescript",
        "markdown": "https://jobs.talent247.co/jobs/senior-agentic-orchestration-engineer-typescript.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-agentic-orchestration-engineer-typescript/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-agentic-orchestration-engineer-typescript/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "f64d9475-8e67-4c8a-9bab-afa20f205c9c",
      "slug": "senior-research-scientist-flow-matching-visual",
      "title": "Senior Research Scientist, Flow-Matching Visual Foundation Models",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Freiburg / Berlin, Germany (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 125000,
        "max": 180000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€125k – €180k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Scientist, Flow-Matching Visual Foundation Models (AI Research) — Freiburg / Berlin, Germany · Hybrid · €125k – €180k. THE CLIENT &…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally acclaimed frontier visual intelligence laboratory headquartered in Freiburg, Germany (with a second hub in Berlin), founded by the original researchers behind latent diffusion and high-resolution flow-matching transformers. Backed by tier-one Silicon Valley and European investors at unicorn valuation, they train state-of-the-art foundation models for image synthesis, context-aware visual editing, typography rendering, and generative video.\n\nThey have partnered with Talent247 to appoint a Senior Research Scientist, Flow-Matching Visual Foundation Models in Freiburg (or Berlin hybrid). Joining a compact, exceptionally high-density team of world-renowned computer vision researchers, you will own core architectural and pre-training breakthroughs for their next generation of multimodal visual and video foundation models.\n\nKEY RESPONSIBILITIES\n• Conceive, design, and pre-train multi-billion-parameter rectified-flow and diffusion-transformer (DiT / MMDiT) architectures unifying text-to-image, instruction-guided editing, in-context visual reasoning, and temporal video synthesis.\n• Innovate on continuous and discrete visual/spatiotemporal tokenizers (autoencoders) maximizing perceptual reconstruction fidelity, compression ratio, and latent-space trainability.\n• Design post-training alignment recipes—spanning DPO, differentiable reward backpropagation, and RLHF for aesthetics, physical plausibility, text rendering, and multi-subject compositionality.\n• Scale stable mixed-precision pre-training runs across thousands of GPUs, diagnosing gradient spikes, representation collapse, and multi-resolution curriculum tradeoffs.\n• Curate and filter billion-scale image-video-caption training mixtures using synthetic VLM recaptioning and semantic deduplication.\n• Publish landmark technical reports and open-weight milestones while collaborating directly with inference engineers to bring frontier models to millions of users.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers enjoy massive dedicated H100/B200 compute per scientist, zero bureaucratic overhead, and a clean PyTorch/CUDA distributed training codebase. Offered hybrid in Freiburg or Berlin, compensation includes €125,000–€180,000 EUR base salary plus premier frontier-lab equity, 30 days paid leave, and full relocation/visa support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Computer Vision, Machine Learning, or a related field, or an equivalent track record of seminal industrial research in generative visual modeling.\n• First-author publications at premier venues (CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, or SIGGRAPH) in diffusion models, flow matching, autoregressive visual generation, or neural tokenizers.\n• Expert hands-on mastery of PyTorch and large-scale distributed multi-node GPU pre-training (FSDP, tensor/sequence parallelism, BF16/FP8).\n• Strong intuition for scaling laws, multimodal text-vision conditioning, VAE/GAN latent representations, and perceptual evaluation metrics.\n• Self-directed builder mentality—comfortable writing training loops, data loaders, and custom losses directly from scratch.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience co-leading a frontier open-weight or commercial text-to-image / text-to-video foundation model release.\n• Familiarity with custom CUDA/Triton attention kernels and step-distillation techniques.",
      "updated_at": "2026-11-15T17:24:52.307Z",
      "created_at": "2026-11-15T16:54:59.437Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-scientist-flow-matching-visual",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-scientist-flow-matching-visual.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-flow-matching-visual/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-flow-matching-visual/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "c1eddf54-a12c-44c7-82b3-b33acfbd516b",
      "slug": "senior-diffusion-distillation-gpu-inference",
      "title": "Senior Diffusion Distillation & GPU Inference Engineer",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Berlin, Germany (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 105000,
        "max": 150000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€105k – €150k",
        "subtitle": "annual package"
      },
      "summary": "Senior Diffusion Distillation & GPU Inference Engineer (AI Infrastructure) — Berlin, Germany · Hybrid · €105k – €150k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of the most celebrated frontier generative media laboratories in the world, founded in Germany by the core scientific pioneers behind modern latent diffusion and rectified-flow visual generation. Operating out of Berlin and Freiburg, their open-weight and API visual foundation models power global creative suites, design platforms, e-commerce studios, and consumer AI products at tens of millions of daily generations.\n\nThey have retained Talent247 to hire a Senior Diffusion Distillation & GPU Inference Engineer in Berlin. Working shoulder-to-shoulder with the researchers inventing next-generation image, editing, and video Diffusion Transformers (DiTs), you will compress 50-step multi-billion-parameter flow-matching models into sub-second, interactive 1-to-4-step production engines running at world-record GPU efficiency.\n\nKEY RESPONSIBILITIES\n• Research, implement, and scale step-distillation and architectural compression recipes—including adversarial diffusion distillation, consistency/rectified-flow trajectory distillation, and width/depth pruning—for 10B+ parameter image and video DiTs.\n• Author and tune custom CUDA, CUTLASS, and Triton kernels for quantized attention (FP8, NVFP4, INT4), fused AdaLN modulation, 3D spatiotemporal RoPE, and high-resolution VAE decoding.\n• Profile end-to-end serving stacks with Nsight Systems and Nsight Compute across H100, H200, and Blackwell B200 GPUs to eliminate host-overhead, memory-bandwidth stalls, and multi-GPU communication bottlenecks.\n• Implement sequence/context parallelism (USP/Ring Attention) and dynamic resolution batching to serve megapixel image editing and high-frame-rate video generation concurrently.\n• Build perceptual and human-preference evaluation pipelines ensuring distilled and quantized checkpoints preserve typography accuracy, photorealistic detail, and prompt adherence.\n• Release industry-leading open-source reference inference implementations alongside commercial API rollouts.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack runs on PyTorch, C++20, CUDA/CUTLASS, Triton, torch.compile, TensorRT, and dedicated multi-hundred-H100/B200 clusters. Working hybrid from Berlin (or Freiburg), our client offers €105,000–€150,000 EUR base salary plus high-upside frontier-lab equity, 30 days vacation, and full German visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in ML performance engineering, GPU kernel optimization, or generative computer vision model compression.\n• Deep proficiency in PyTorch, C++, and CUDA or Triton kernel programming, with proven speedups on large transformer or diffusion architectures.\n• Hands-on experience with diffusion/flow-matching distillation techniques (consistency models, adversarial/distribution-matching distillation, CFG distillation) or low-bit quantization (FP8/INT4/NVFP4).\n• Strong hardware-level intuition for Tensor Cores, shared memory tiling, CUDA graphs, `torch.compile`/Inductor, and multi-GPU NCCL collectives.\n• Track record shipping production-grade inference code under high-throughput commercial API traffic.\n\nPREFERRED / BONUS EXPERIENCE\n• Publications at CVPR, ICCV, ECCV, NeurIPS, or MLSys, or recognized open-source contributions to diffusers, sglang, TensorRT, FlashAttention, or Triton.\n• Experience optimizing 3D causal video VAEs and real-time streaming video generation.",
      "updated_at": "2026-11-15T17:24:42.558Z",
      "created_at": "2026-11-15T16:55:02.770Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-diffusion-distillation-gpu-inference",
        "markdown": "https://jobs.talent247.co/jobs/senior-diffusion-distillation-gpu-inference.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-diffusion-distillation-gpu-inference/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-diffusion-distillation-gpu-inference/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "b0609582-24c7-4d3e-831c-144eb3b2d22a",
      "slug": "staff-edge-ai-multi-sensor-fusion-engineer",
      "title": "Staff Edge AI & Multi-Sensor Fusion Engineer",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "Munich, Germany (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 115000,
        "max": 155000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€115k – €155k",
        "subtitle": "annual package"
      },
      "summary": "Staff Edge AI & Multi-Sensor Fusion Engineer (Robotics & Embodied AI) — Munich, Germany · On-site · €115k – €155k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Europe's premier defence and real-time edge AI unicorn headquartered in Munich, partnering with European defence ministries and aerospace primes to equip modern platforms with real-time sensor fusion, cognitive electronic protection, and onboard mission intelligence. Their systems fuse high-bandwidth radar, electro-optical/infrared (EO/IR), acoustic, and RF streams directly on ruggedized tactical edge compute with zero cloud connectivity.\n\nThey have engaged Talent247 to appoint a Staff Edge AI & Multi-Sensor Fusion Engineer onsite in Munich. Acting as a principal technical lead across ML perception and embedded systems, you will architect spatio-temporal multi-sensor fusion models and the ultra-deterministic C++/CUDA/TensorRT onboard pipelines that run them at millisecond latency in high-consequence environments.\n\nKEY RESPONSIBILITIES\n• Lead the end-to-end architecture of multi-modal deep learning perception and tracking models fusing asynchronous radar, EO/IR video, passive RF, and inertial streams under severe noise, clutter, and spoofing.\n• Architect deterministic, zero-allocation C++20/Rust and TensorRT/CUDA inference pipelines targeting embedded tactical GPU/SoC hardware under strict SWaP (size, weight, and power) and thermal budgets.\n• Pioneer quantization-aware training (INT8/FP8), structured pruning, custom CUDA/Triton fusion kernels, and sensor-level early-exit architectures to maximize onboard frame rates.\n• Validate model and runtime behavior inside hardware-in-the-loop (HIL) rigs and live field trials, diagnosing subtle timing skew, sensor-calibration drift, and numerical precision edge cases.\n• Define engineering standards for high-assurance onboard AI software, mentoring senior ML and embedded engineers across Munich and European hubs.\n• Serve as a senior technical counterpart to external sensor/platform OEMs and programme systems engineers.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir environment spans C++20, Rust, Python, PyTorch, CUDA, TensorRT, real-time Linux/QNX, DDS/custom low-latency middleware, and dedicated multi-node training clusters. Based onsite at their flagship Munich R&D headquarters, our client offers €115,000–€155,000 EUR base salary plus high-value equity, relocation support, and comprehensive German benefits. EU/NATO security-clearance eligibility is mandatory.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of hands-on engineering experience spanning computer vision / signal-processing deep learning and high-performance embedded C++/GPU deployment.\n• Mastery of modern C++ (C++17/20) and Python/PyTorch, coupled with deep practical expertise in CUDA, TensorRT, and NVIDIA Jetson/Orin/Thor or comparable tactical edge accelerators.\n• Proven track record designing multi-sensor fusion, 3D multi-object tracking, radar/RF deep learning, or real-time spatio-temporal perception systems.\n• Strong systems debugging command across real-time Linux scheduling, DMA/zero-copy camera/sensor buffers, and hardware-in-the-loop validation.\n• Eligibility for German / EU / NATO national security clearance.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with aerospace/defence software assurance standards, electronic warfare / radar signal processing, or safety-critical autonomous systems.\n• Proficiency in Rust for high-assurance systems components.",
      "updated_at": "2026-11-15T17:24:31.305Z",
      "created_at": "2026-11-15T16:55:06.266Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-edge-ai-multi-sensor-fusion-engineer",
        "markdown": "https://jobs.talent247.co/jobs/staff-edge-ai-multi-sensor-fusion-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-edge-ai-multi-sensor-fusion-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-edge-ai-multi-sensor-fusion-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "134e6e26-fd4c-4751-9e8c-2a472f21e5d5",
      "slug": "senior-reinforcement-learning-engineer-autonomous",
      "title": "Senior Reinforcement Learning Engineer, Autonomous Systems",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Berlin, Germany (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 100000,
        "max": 140000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€100k – €140k",
        "subtitle": "annual package"
      },
      "summary": "Senior Reinforcement Learning Engineer, Autonomous Systems (Machine Learning) — Berlin, Germany · Hybrid · €100k – €140k. THE CLIENT & MANDATE Our client…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Europe's highest-valued defence and autonomous-systems artificial intelligence company, operating major R&D centers in Berlin, Munich, London, and Paris. They build software-defined autonomy, electronic-warfare intelligence, and multi-domain coordination platforms engineered to protect democratic nations, deploying advanced machine learning onto airborne, maritime, and ground platforms operating in GPS-denied and communications-contested environments.\n\nThey have retained Talent247 to recruit a Senior Reinforcement Learning Engineer, Autonomous Systems in Berlin. Embedded within their Autonomy & Mission Planning R&D division, you will train single- and multi-agent reinforcement learning (MARL) policies inside massively parallel physics simulations and transfer those learned behaviors safely onto physical autonomous platforms.\n\nKEY RESPONSIBILITIES\n• Design, train, and scale deep reinforcement learning and imitation-learning policies (PPO, SAC, Dreamer/world-model RL, offline-to-online RL, MARL) for tactical maneuvering, sensor tasking, and decentralized multi-vehicle coordination.\n• Architect GPU-accelerated, high-throughput physics and RF/sensor simulation environments capable of generating hundreds of millions of diverse, adversarial training steps per hour.\n• Develop domain-randomization, system-identification, and safety-shielding (control barrier functions / constrained RL) techniques that bridge the sim-to-real gap reliably during live flight and sea trials.\n• Build rigorous red-vs-blue adversarial evaluation tournaments and deterministic scenario suites quantifying policy robustness under sensor degradation, jamming, and partial observability.\n• Collaborate closely with guidance-navigation-control (GNC) and embedded C++/Rust engineers to compile compact policy networks for deterministic onboard execution.\n• Participate in field-test analysis loops, translating real-world telemetry anomalies directly into improved simulation fidelity and reward shaping.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack combines Python, PyTorch, JAX, C++20, Rust, CUDA-accelerated simulation engines, Ray/Kubernetes GPU clusters, and hardware-in-the-loop testbeds. Working hybrid in central Berlin, our client offers €100,000–€140,000 EUR base salary plus substantial unicorn equity options, 30 days holiday, and relocation assistance. Note: candidates must hold citizenship eligible for EU/NATO security clearance.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of research or engineering experience in deep reinforcement learning, autonomous robotics, optimal control, or game-theoretic multi-agent systems.\n• Expert proficiency in PyTorch or JAX for distributed RL training, alongside strong C++ or Rust skills for simulation and runtime integration.\n• Proven track record training policies in complex continuous-control 3D simulation environments and validating sim-to-real transfer or hardware-in-the-loop performance.\n• Solid mathematical grounding in MDPs/POMDPs, multi-agent RL (MARL), trajectory optimization, safety-constrained control, or world models.\n• Eligibility to obtain German / EU / NATO security clearance based on national defence regulatory requirements.\n\nPREFERRED / BONUS EXPERIENCE\n• PhD in Robotics, Reinforcement Learning, Aerospace/Control Engineering, or Computer Science with publications at CoRL, RSS, ICRA, NeurIPS, or ICML.\n• Prior experience with UAV/AUV flight dynamics, sensor management, or GPU-resident physics simulators (Isaac Lab, Brax, MuJoCo MJX, or custom CUDA sims).",
      "updated_at": "2026-11-15T17:24:21.235Z",
      "created_at": "2026-11-15T16:55:09.800Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-reinforcement-learning-engineer-autonomous",
        "markdown": "https://jobs.talent247.co/jobs/senior-reinforcement-learning-engineer-autonomous.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-reinforcement-learning-engineer-autonomous/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-reinforcement-learning-engineer-autonomous/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "06be68f5-2a3c-4bec-9c6f-0c2133b0e077",
      "slug": "senior-mlops-enterprise-llm-serving-platform",
      "title": "Senior MLOps & Enterprise LLM Serving Platform Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Berlin / Cologne, Germany (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 92000,
        "max": 125000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€92k – €125k",
        "subtitle": "annual package"
      },
      "summary": "Senior MLOps & Enterprise LLM Serving Platform Engineer (MLOps & Platform) — Berlin / Cologne, Germany · Hybrid · €92k – €125k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar German enterprise AI unicorn with major engineering centers in Berlin and Cologne, trusted by more than 100,000 businesses worldwide for mission-critical neural language intelligence and autonomous cross-lingual enterprise workflows. Unlike application wrappers renting third-party APIs, they train proprietary foundation models on dedicated multi-thousand-GPU supercomputing clusters and serve billions of low-latency enterprise requests daily under strict European data-privacy guarantees.\n\nThey have engaged Talent247 to hire a Senior MLOps & Enterprise LLM Serving Platform Engineer in Berlin. Joining their Inference & ML Platform division, you will own the internal model deployment control plane, automated regression/canary pipelines, and dynamic GPU cluster scheduling powering real-time translation, writing, voice, and enterprise agent workloads.\n\nKEY RESPONSIBILITIES\n• Architect, scale, and operate Kubernetes-native LLM and multimodal model serving platforms running custom C++/CUDA, Triton, and vLLM/TensorRT-LLM engines across bare-metal and cloud GPU estates.\n• Build zero-downtime model rollout workflows featuring shadow traffic mirroring, automated quality/latency canary gates, instant rollback, and cryptographic artifact provenance.\n• Design intelligent request routing, prefix-cache-aware load balancing, and multi-tier autoscaling policies that hold sub-200ms interactive SLAs while maximizing fleet GPU utilization.\n• Instrument deep GPU and inference observability (TTFT, TPOT, KV-cache pressure, SM occupancy, NVLink throughput) using Prometheus, OpenTelemetry, DCGM, and Grafana.\n• Partner directly with research scientists and kernel engineers to streamline the path from multi-node training checkpoint to quantized production endpoint.\n• Drive GitOps automation, disaster-recovery readiness, and strict zero-retention enterprise compliance across European and North American serving regions.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir platform stack spans Kubernetes, Go, Python, Rust, Triton Inference Server, TensorRT-LLM, vLLM, Envoy, ArgoCD, and dedicated H100/B200 superclusters. Offered hybrid in Berlin (or Cologne), compensation includes €92,000–€125,000 EUR base salary, equity participation, 30 days annual leave, German relocation/Blue Card visa support, and professional development budgets.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of MLOps, ML platform, or production infrastructure engineering experience operating high-throughput distributed services on Kubernetes.\n• Direct production experience serving large transformer models using vLLM, TensorRT-LLM, Triton Inference Server, SGLang, or custom C++/CUDA runtimes.\n• Strong programming skills in Python and Go (or Rust) for writing Kubernetes operators, deployment controllers, and performance benchmarking tools.\n• Deep operational familiarity with Linux networking, NVIDIA GPU Operator/MIG/DCGM telemetry, Envoy/gRPC load balancing, and GitOps (ArgoCD/Flux, Terraform).\n• Proven ability to diagnose complex tail-latency and throughput regressions across model serving stacks.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience operating bare-metal HPC/Slurm or multi-thousand-GPU data center environments alongside cloud Kubernetes clusters.\n• Working knowledge of model quantization (FP8/INT4), speculative decoding, or disaggregated prefill/decode serving architectures.",
      "updated_at": "2026-11-15T17:24:09.782Z",
      "created_at": "2026-11-15T16:55:13.129Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform",
        "markdown": "https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "dab7f703-4047-402b-8c3b-60c2be025603",
      "slug": "senior-adversarial-ml-agent-runtime-security",
      "title": "Senior Adversarial ML & Agent Runtime Security Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Zurich, Switzerland (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 145000,
        "max": 190000,
        "currency": "CHF",
        "frequency": "yearly",
        "formatted": "CHF 145k – CHF 190k",
        "subtitle": "annual package"
      },
      "summary": "Senior Adversarial ML & Agent Runtime Security Engineer (Machine Learning) — Zurich, Switzerland · Hybrid · CHF 145k – CHF 190k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an acclaimed AI security pioneer headquartered in Zurich, spun out of Switzerland's world-renowned machine learning and cybersecurity ecosystem. As enterprises transition from passive chatbots to autonomous agents granted write access across internal databases, codebases, and APIs, our client provides the real-time runtime defense layer and continuous red-teaming engine keeping those agents safe against indirect prompt injection, jailbreaks, and tool hijacking.\n\nThey have retained Talent247 to appoint a Senior Adversarial ML & Agent Runtime Security Engineer at their Zurich headquarters. Sitting on their Threat Research & Guardrail Modeling team, you will co-develop both offensive automated red-teaming agents that uncover novel foundation-model vulnerabilities and sub-15ms defensive classifier architectures deployed inline across global enterprise traffic.\n\nKEY RESPONSIBILITIES\n• Research, train, and distil ultra-fast multilingual transformer and encoder/decoder guardrail models detecting indirect prompt injection, toxic content, PII leakage, and unauthorized tool-call trajectories under strict <15ms p99 latency SLAs.\n• Build evolutionary and RL-driven offensive red-teaming agents that automatically probe newly released closed- and open-weight LLMs and multi-agent harnesses for novel jailbreaks and cross-context exploits.\n• Curate massive multilingual adversarial datasets leveraging red-team telemetry from hundreds of thousands of attack simulations alongside verified enterprise edge cases.\n• Minimize false-positive friction on benign developer, code, and security-analyst prompts through hard-negative synthesis, threshold calibration, and domain-adaptive fine-tuning.\n• Collaborate closely with Rust runtime engineers to quantize (INT8/FP8) and compile compact classifier heads for both CPU sidecars and GPU clusters.\n• Publish threat intelligence advisories, open benchmarks, and peer-reviewed AI security research representing our client across the European and global safety community.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack includes Python, PyTorch, ONNX/TensorRT distillation, vLLM red-team fleets, Rust inference gateways, and multi-GPU training clusters. Working hybrid in central Zurich, our client offers CHF 145,000–CHF 190,000 base salary plus generous equity, Swiss pension and benefits, and full Swiss work-permit/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in machine learning engineering, NLP, or adversarial AI security research with production model deployment ownership.\n• Deep proficiency in PyTorch, encoder/decoder transformer architectures, knowledge distillation, quantization-aware training, and synthetic adversarial data generation.\n• Practical understanding of LLM and agent security threat vectors: direct/indirect prompt injection, RAG poisoning, MCP/tool-use hijacking, and multi-turn jailbreaks.\n• Strong experimental rigor around extreme precision-at-high-recall tradeoffs, ROC calibration, and out-of-distribution robustness across dozens of languages.\n• Fluent English communication and collaborative engineering habits across ML research and systems teams.\n\nPREFERRED / BONUS EXPERIENCE\n• PhD or MSc in Machine Learning, Cybersecurity, or NLP, or publications/CTF track record in adversarial ML and LLM red-teaming.\n• Experience optimizing sub-20ms transformer inference on CPU (AVX-512/AMX, ONNX Runtime) or GPU (TensorRT/Triton).",
      "updated_at": "2026-11-15T17:23:58.221Z",
      "created_at": "2026-11-15T16:55:16.723Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-adversarial-ml-agent-runtime-security",
        "markdown": "https://jobs.talent247.co/jobs/senior-adversarial-ml-agent-runtime-security.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-adversarial-ml-agent-runtime-security/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-adversarial-ml-agent-runtime-security/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "b2cabccb-5e70-49db-b9cf-99b9ebfb8d6f",
      "slug": "senior-rust-streaming-gateway-inline-guardrails",
      "title": "Senior Rust Streaming Gateway & Inline Guardrails Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Warsaw, Poland (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 80000,
        "max": 110000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€80k – €110k",
        "subtitle": "annual package"
      },
      "summary": "Senior Rust Streaming Gateway & Inline Guardrails Engineer (Applied AI) — Warsaw, Poland · Hybrid · €80k – €110k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an ETH-ecosystem AI security leader with engineering hubs in Zurich and Warsaw, safeguarding live enterprise LLM applications and autonomous multi-agent workflows for global banks, healthcare networks, and Fortune 500 enterprises. Their inline runtime security platform inspects billions of prompts, retrieved context chunks, and tool invocations monthly to block prompt injection, sensitive data exfiltration, and unauthorized agent actions in real time.\n\nThey have partnered with Talent247 to hire a Senior Rust Streaming Gateway & Inline Guardrails Engineer in Warsaw. Joining their Core Runtime & Systems Engineering squad, you will architect ultra-low-latency asynchronous proxy gateways and embedded neural inference engines capable of enforcing deterministic and ML-based security policies within single-digit millisecond budgets.\n\nKEY RESPONSIBILITIES\n• Design, build, and optimize high-concurrency Rust streaming reverse proxies (Tokio, Hyper, Envoy ext_proc) inspecting SSE token streams, JSON-RPC/MCP tool calls, and multimodal payloads on the wire.\n• Embed quantized small-transformer guardrail classifiers directly inside the gateway process via ONNX Runtime, Candle, or TensorRT to hold p99 inline inspection latency below 10ms.\n• Architect zero-data-retention memory handling, PII/secret redaction state machines, and cryptographic audit-log streaming for regulated financial and healthcare customers.\n• Package self-contained, high-availability Kubernetes Helm deployments and WebAssembly/sidecar plugins for seamless customer VPC and on-premise rollout.\n• Build chaos-testing and high-RPS load harnesses simulating millions of concurrent streaming agent sessions and adversarial payload bursts.\n• Partner closely with adversarial ML researchers in Zurich to ship new threat-detection models from research checkpoint to production gateway within hours.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir systems stack centers on Rust, C++, Tokio/Hyper, Envoy, ONNX Runtime, Kubernetes, Terraform, and OpenTelemetry. Working hybrid in central Warsaw, our client offers €80,000–€110,000 EUR base compensation, meaningful equity options, private healthcare, and conference/hardware budgets.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems or high-throughput backend engineering experience, including substantial production ownership in Rust (or modern C++20 with strong Rust fluency).\n• Deep understanding of asynchronous network programming (Tokio, Hyper, gRPC, HTTP/2, SSE streaming), reverse proxies, and Linux network/memory profiling.\n• Hands-on experience embedding low-latency local ML inference (ONNX Runtime, Candle, libtorch, or TensorRT) inside high-QPS production services.\n• Strong operational command of Kubernetes, Helm, multi-region cloud deployments, and distributed tracing/observability (OpenTelemetry, Prometheus).\n• Security-minded engineering discipline around memory safety, tenant isolation, and zero-trust API design.\n\nPREFERRED / BONUS EXPERIENCE\n• Background building WAFs, API gateways, service meshes (Envoy/Istio), DLP engines, or LLM guardrail/MCP proxy tooling.\n• Contributions to open-source Rust networking or ML inference crates.",
      "updated_at": "2026-11-15T17:23:48.453Z",
      "created_at": "2026-11-15T16:55:20.597Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-rust-streaming-gateway-inline-guardrails",
        "markdown": "https://jobs.talent247.co/jobs/senior-rust-streaming-gateway-inline-guardrails.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-rust-streaming-gateway-inline-guardrails/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-rust-streaming-gateway-inline-guardrails/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "4619b4b5-d017-4b09-90e9-ff4a6b171f4e",
      "slug": "staff-machine-learning-engineer-multilingual",
      "title": "Staff Machine Learning Engineer, Multilingual Embeddings & Reranking",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Toronto, ON, Canada (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 195000,
        "max": 270000,
        "currency": "CAD",
        "frequency": "yearly",
        "formatted": "CA$195k – CA$270k",
        "subtitle": "annual package"
      },
      "summary": "Staff Machine Learning Engineer, Multilingual Embeddings & Reranking (Machine Learning) — Toronto, ON, Canada · Hybrid · CA$195k – CA$270k. THE CLIENT &…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a multi-billion-dollar enterprise AI foundation model company headquartered in Toronto whose multilingual language models, dense/sparse embeddings, and neural rerankers power production search and agentic RAG workloads across the Global 2000. Unlike general consumer chat labs, their entire research roadmap is dedicated to enterprise-grade accuracy, cross-lingual retrieval across 100+ languages, verifiable citations, and high-efficiency enterprise deployment.\n\nThey have retained Talent247 to appoint a Staff Machine Learning Engineer, Multilingual Embeddings & Reranking in Toronto. Sitting as a technical pillar within their Retrieval & Representation Learning group, you will lead the architecture, contrastive pre-training, and post-training distillation of next-generation text and multimodal embedding and cross-encoder reranking models used by millions of enterprise knowledge workers daily.\n\nKEY RESPONSIBILITIES\n• Set technical direction and own end-to-end training recipes for frontier multilingual text and multimodal (slide, table, chart, PDF) embedding and late-interaction/cross-encoder reranking models.\n• Design large-scale contrastive pre-training, synthetic query-document generation, LLM-guided hard-negative mining, and distillation pipelines across 100+ high- and low-resource languages.\n• Pioneer Matryoshka representation learning, binary/int8 vector quantization, and asymmetric query-document encoding architectures that slash customer vector-index RAM costs by 8x–32x without recall degradation.\n• Partner with enterprise agent and RAG research teams to co-train retrieval models alongside reasoning LLMs for multi-hop query decomposition and tool discovery.\n• Build gold-standard enterprise retrieval benchmarks spanning finance, legal, code, technical manuals, and noisy OCR documents, driving measurable gains beyond public academic leaderboards.\n• Mentor senior ML engineers and collaborate with inference teams to ship sub-15ms p99 reranking endpoints in production.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir modeling stack runs on JAX and PyTorch across large multi-slice TPU and H100 GPU clusters, backed by petabyte-scale multilingual data curation pipelines. Based hybrid in downtown Toronto, this Staff mandate offers CA$195,000–CA$270,000 base salary plus substantial equity, premier healthcare benefits, and full relocation/immigration support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of hands-on experience in machine learning engineering or applied research, with deep specialization in neural information retrieval, representation learning, or foundation model training.\n• Expert proficiency in PyTorch or JAX, distributed multi-node training, contrastive loss formulations, knowledge distillation, and large-scale synthetic data curation.\n• Proven track record shipping state-of-the-art dense/sparse embedding models, cross-encoder rerankers, or multimodal retrieval systems at scale.\n• Deep familiarity with ANN vector search internals (HNSW, DiskANN, PQ, binary/scalar quantization) and storage-vs-recall tradeoffs in enterprise RAG.\n• Track record of technical leadership, experimental rigor, and mentoring engineers in a fast-paced AI lab environment.\n\nPREFERRED / BONUS EXPERIENCE\n• Publications at SIGIR, ACL, EMNLP, NeurIPS, ICML, or ICLR, or top-ranked releases on major retrieval/embedding benchmarks (MTEB, BEIR, MIRACL, ViDoRe).\n• Experience with multimodal vision-language document retrieval or agentic multi-hop search loops.",
      "updated_at": "2026-11-15T17:23:38.107Z",
      "created_at": "2026-11-15T16:55:42.373Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-machine-learning-engineer-multilingual",
        "markdown": "https://jobs.talent247.co/jobs/staff-machine-learning-engineer-multilingual.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-machine-learning-engineer-multilingual/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-machine-learning-engineer-multilingual/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "a364cdd1-63ed-46ea-84fb-afb347fcbf9e",
      "slug": "senior-ml-engineer-generative-protein-design",
      "title": "Senior ML Engineer, Generative Protein Design Models",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Amsterdam, Netherlands (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 100000,
        "max": 135000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€100k – €135k",
        "subtitle": "annual package"
      },
      "summary": "Senior ML Engineer, Generative Protein Design Models (Machine Learning) — Amsterdam, Netherlands · Hybrid · €100k – €135k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a fast-growing European AI-for-protein-engineering company headquartered in Amsterdam with a twin R&D hub in Zurich, trusted by leading global biopharma and industrial biotechnology teams to design custom therapeutic antibodies, vaccines, and sustainable enzymes. Their platform pairs proprietary sequence-and-structure generative models with an intuitive scientist studio and rapid wet-lab feedback loops.\n\nThey have retained Talent247 to hire a Senior Machine Learning Engineer, Generative Protein Design Models in Amsterdam. Embedded in an interdisciplinary squad of ML researchers, software engineers, and protein engineers, you will own the production training, fine-tuning, and multi-objective sequence optimization pipelines that deliver wet-lab-ready candidate libraries directly to biopharma partners.\n\nKEY RESPONSIBILITIES\n• Build, scale, and productionize fine-tuning, preference-alignment, and active-learning pipelines over large protein language models and structure-conditioned generative architectures.\n• Engineer multi-objective sequence design workflows balancing competing therapeutic properties (affinity, developability, solubility, humanization, expression) across iterative experimental rounds.\n• Design automated benchmark suites and prospective backtesting harnesses across internal wet-lab datasets and public protein engineering benchmarks (ProteinGym, FLIP, antibody binding panels).\n• Optimize distributed PyTorch training, dataset sharding, mixed-precision throughput, and low-latency GPU inference so wet-lab scientists can interactively score and generate thousands of variants in real time.\n• Partner with platform and product engineers to expose calibrated model uncertainty scores, interpretability attributions, and zero-data-leakage tenant fine-tuning inside the customer platform.\n• Contribute clean, well-tested production ML code, architectural reviews, and technical mentorship across the engineering team.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack combines Python, PyTorch, Ray, Kubernetes on GCP/AWS, FastAPI/Rust services, and modern experiment-tracking/lineage tooling. Working hybrid in central Amsterdam (eligible for the Dutch 30% ruling for qualifying international hires), our client offers €100,000–€135,000 EUR base salary, generous stock options, Dutch pension, and full visa/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience as a Machine Learning Engineer building, training, and deploying deep learning sequence, graph, or generative models in production.\n• High proficiency in Python and PyTorch, including distributed multi-GPU training, parameter-efficient fine-tuning (LoRA/adapters), and reproducible ML pipeline design.\n• Hands-on experience with biological sequence models (ESM, ProtTrans, Evo, ProGen) or comparable domain-specific transformer/diffusion architectures, alongside strong willingness to go deep on protein biology.\n• Solid grasp of Bayesian optimization, active learning, uncertainty calibration, or reinforcement learning under noisy, small-data experimental regimes.\n• Strong software engineering craft: modular architecture, automated testing, CI/CD, and clear communication with non-ML domain scientists.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior experience in computational protein design, antibody engineering, structural prediction (AlphaFold/Boltz/ESMFold ecosystems), or cheminformatics.\n• Experience scaling Ray/Kubernetes GPU batch inference and multi-tenant per-customer model fine-tuning.",
      "updated_at": "2026-11-15T17:22:44.397Z",
      "created_at": "2026-11-15T16:55:23.617Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-ml-engineer-generative-protein-design",
        "markdown": "https://jobs.talent247.co/jobs/senior-ml-engineer-generative-protein-design.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-generative-protein-design/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-ml-engineer-generative-protein-design/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3540a3dd-3b36-43bd-a649-a0c3da26d0c6",
      "slug": "research-scientist-biological-sequence-foundation",
      "title": "Research Scientist, Biological Sequence Foundation Models",
      "department": "AI Research",
      "department_code": "RES",
      "location": "Zurich, Switzerland (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 150000,
        "max": 200000,
        "currency": "CHF",
        "frequency": "yearly",
        "formatted": "CHF 150k – CHF 200k",
        "subtitle": "annual package"
      },
      "summary": "Research Scientist, Biological Sequence Foundation Models (AI Research) — Zurich, Switzerland · Hybrid · CHF 150k – CHF 200k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a premier European generative AI-for-biology company operating tightly integrated dry-lab ML research and automated wet-lab facilities across Zurich and Amsterdam. Backed by top trans-Atlantic venture funds, they build foundation models for protein programming that allow pharmaceutical and industrial-biotech partners to simultaneously optimize binding affinity, thermostability, expression yield, and immunogenicity across far fewer experimental wet-lab rounds.\n\nThey have partnered with Talent247 to appoint a Research Scientist, Biological Sequence Foundation Models at their Zurich research center. Working closely with computational biologists, ML engineers, and wet-lab protein scientists, you will design and pre-train next-generation evolutionary, structural, and assay-conditioned foundation models that learn the underlying grammar of functional biomolecules.\n\nKEY RESPONSIBILITIES\n• Formulate, pre-train, and scale novel autoregressive, masked-diffusion, and flow-matching foundation models over billions of evolutionary protein/nucleotide sequences, 3D structural tokens, and biophysical assay readouts.\n• Develop sample-efficient post-training,test-time guidance, and Bayesian/RL active-learning methods that adapt foundation representations to sparse, noisy multi-parameter wet-lab experimental data.\n• Design calibrated uncertainty-estimation and out-of-distribution generalization frameworks so wet-lab synthesis budgets focus on high-probability, diverse functional variants.\n• Collaborate directly with internal wet-lab biologists to design prospective experimental validation campaigns that rigorously test new modeling hypotheses in physical assays within weeks.\n• Scale clean, modular JAX or PyTorch pre-training code across multi-node H100 clusters, optimizing tokenization, data mixtures, and long-context sequence attention.\n• Publish impactful methodological advances at leading ML and computational biology venues and mentor fellow researchers.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir research platform runs on PyTorch/JAX, multi-node GPU clusters, custom biological sequence/structure data lakes, and automated high-throughput wet-lab pipelines. Based hybrid in central Zurich, our client offers CHF 150,000–CHF 200,000 base salary plus meaningful equity, Swiss pension and social benefits, and full Swiss work-permit/relocation assistance.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Machine Learning, Computational Biology, Bioinformatics, Biophysics, or a related quantitative discipline (or equivalent frontier research experience).\n• Strong publication track record at premier venues such as NeurIPS, ICML, ICLR, RECOMB, ISMB, Nature Methods, or Nature Machine Intelligence.\n• Deep hands-on expertise pre-training or post-training large transformer, state-space, or discrete diffusion foundation models in PyTorch or JAX.\n• Solid understanding of protein language models, 3D structural representations, multiple-sequence alignments (MSAs), or active learning in experimental design.\n• Ability to bridge mathematical modeling rigor with clean distributed engineering and cross-disciplinary communication with wet-lab scientists.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct experience validating ML-generated antibodies, enzymes, or peptides in closed-loop wet-lab design-build-test-learn cycles.\n• Experience with multi-objective optimization, GFlowNets, reinforcement learning for molecular design, or custom CUDA/Triton attention kernels.",
      "updated_at": "2026-11-15T17:22:44.075Z",
      "created_at": "2026-11-15T16:55:27.488Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/research-scientist-biological-sequence-foundation",
        "markdown": "https://jobs.talent247.co/jobs/research-scientist-biological-sequence-foundation.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/research-scientist-biological-sequence-foundation/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/research-scientist-biological-sequence-foundation/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "968c13f9-7fb7-4bb8-9483-550879a237bc",
      "slug": "staff-agentic-code-generation-cloud-sandbox",
      "title": "Staff Agentic Code Generation & Cloud Sandbox Engineer",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Stockholm, Sweden (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 105000,
        "max": 145000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€105k – €145k",
        "subtitle": "annual package"
      },
      "summary": "Staff Agentic Code Generation & Cloud Sandbox Engineer (Applied AI) — Stockholm, Sweden · On-site · €105k – €145k. THE CLIENT & MANDATE Our client is a…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a category-defining Series C AI unicorn headquartered in central Stockholm, building the autonomous software-engineering platform used by millions of creators and enterprise product teams worldwide to chat full-stack applications into production. Behind the conversational interface sits a sophisticated agentic runtime orchestrating multi-file AST edits, ephemeral cloud microVMs, live browser DOM/console telemetry, and managed Postgres backends.\n\nThey have retained Talent247 to recruit a Staff Agentic Code Generation & Cloud Sandbox Engineer onsite in Stockholm. As a senior technical leader on their core Agent & Runtime Engineering team, you will architect both the multi-step planning harness and the sub-second isolated cloud execution environments that allow autonomous coding agents to build, inspect, test, and self-heal complex applications reliably.\n\nKEY RESPONSIBILITIES\n• Architect and evolve the core agentic loop—spanning context-window compaction, repository AST indexing, deterministic tool execution, parallel sub-agent delegation, and automated compile/runtime self-healing.\n• Design and scale ultra-fast ephemeral cloud sandboxes (microVMs/containers) capable of booting full-stack dev servers, running migrations, and streaming live hot-reload previews in hundreds of milliseconds.\n• Build closed-loop browser verification pipelines that capture DOM layout state, network failures, console stack traces, and visual screenshots so the agent catches and repairs bugs before the user ever sees them.\n• Embed security-by-default guardrails into generated applications, including automated secret detection, database Row-Level Security (RLS) static verification, and auth-flow auditing.\n• Drive latency, reliability, and cost-per-task improvements across multi-model routing, prompt prefix caching, and streaming diff application.\n• Set architectural standards, mentor senior engineers, and partner directly with post-training researchers to co-design tool schemas and training trajectories.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack uses TypeScript, Go, Rust, Python, Firecracker/microVMs, Kubernetes, PostgreSQL, Playwright, and multi-provider frontier + custom fine-tuned LLMs. Based onsite in Stockholm, the package includes €105,000–€145,000 EUR base salary, significant Series C equity, full relocation/work-permit sponsorship, and comprehensive Nordic benefits.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 7+ years of software and distributed systems engineering experience, with proven Staff-level technical leadership in high-growth product or developer-tooling environments.\n• Deep proficiency in TypeScript and at least one systems language (Go or Rust), plus working fluency in Python.\n• Hands-on production experience building autonomous LLM coding/browser agents, AST code-patching engines (tree-sitter, TypeScript Compiler API, LSP), or developer cloud sandboxes.\n• Strong systems grasp of Linux containerization/virtualization (Firecracker, gVisor, Docker, overlay filesystems), reverse proxies, and high-concurrency WebSocket/SSE streaming.\n• Obsession with developer craft, tail-latency reduction, and turning brittle LLM behaviors into deterministic software guarantees.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior authorship of IDE extensions, cloud development environments, CI/CD runners, or static security analysis tools for web applications.\n• Experience instrumenting agent trajectories and evaluation harnesses that feed directly into RL/SFT post-training loops.",
      "updated_at": "2026-11-15T17:22:43.560Z",
      "created_at": "2026-11-15T16:55:31.665Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/staff-agentic-code-generation-cloud-sandbox",
        "markdown": "https://jobs.talent247.co/jobs/staff-agentic-code-generation-cloud-sandbox.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/staff-agentic-code-generation-cloud-sandbox/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/staff-agentic-code-generation-cloud-sandbox/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "fd3909b7-8c32-4c0f-871e-9f18d7a80d70",
      "slug": "ai-research-engineer-code-ui-post-training",
      "title": "AI Research Engineer, Code & UI Post-Training",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Stockholm, Sweden (Onsite)",
      "workplace_type": "On-site",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 135000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€95k – €135k",
        "subtitle": "annual package"
      },
      "summary": "AI Research Engineer, Code & UI Post-Training (Machine Learning) — Stockholm, Sweden · On-site · €95k – €135k. THE CLIENT & MANDATE Our client is one of…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is one of Europe's fastest-scaling AI unicorns, headquartered in Stockholm and backed by tier-one US and European venture capital following explosive global adoption. Their agentic software-creation platform enables millions of founders, product builders, and enterprise teams to turn natural-language specifications into deployed, full-stack web applications complete with authentication, databases, and polished interfaces.\n\nThey have engaged Talent247 to hire an AI Research Engineer, Code & UI Post-Training onsite in Stockholm. Sitting at the intersection of model post-training and product execution, you will harness rich build, compiler, runtime, visual-render, and user-acceptance signals from millions of live application generations to train specialized models that write cleaner full-stack code and dramatically higher-craft user interfaces than off-the-shelf foundation models.\n\nKEY RESPONSIBILITIES\n• Own end-to-end post-training recipes (SFT, DPO, and verifiable-reward RL / GRPO) on frontier open-weight code and vision-language models tailored to full-stack TypeScript/React/SQL generation and iterative diff editing.\n• Design automated, multimodal reward functions combining TypeScript compiler/linter checks, Playwright headless browser runtime verification, VLM aesthetic/layout graders, and real user retention telemetry.\n• Train fast, distilled sub-second apply/diff-merging and context-routing models that slash end-to-end generation latency and token spend across millions of daily sessions.\n• Construct contamination-free internal benchmarks covering complex multi-file refactors, database schema migrations, responsive UI design craft, and security-safe auth/RLS policies.\n• Debug large-scale distributed training instabilities, reward hacking behaviors, and data-mixture imbalances across multi-node H100/B200 clusters.\n• Partner tightly with agent-harness and sandbox engineers so model behavior and tool interfaces co-evolve seamlessly.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir R&D stack spans PyTorch, verl/TRL, vLLM/SGLang, TypeScript, Playwright headless sandboxes, ClickHouse telemetry, and multi-node GPU clusters. Based onsite at their high-energy Stockholm headquarters, our client offers €95,000–€135,000 EUR base salary plus high-upside unicorn equity, Swedish relocation and visa sponsorship, catered meals, and top-tier compute access.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 4+ years of hands-on machine learning research engineering experience training, fine-tuning, and evaluating large language or vision-language models in PyTorch.\n• Direct practical experience with modern LLM post-training methodologies, including supervised fine-tuning (SFT), preference optimization (DPO), and reinforcement learning with verifiable rewards (RLVR / GRPO / PPO).\n• Strong software engineering skills in Python (and working familiarity with TypeScript/web stacks to reason deeply about code-generation failure modes).\n• Track record designing automated evaluation harnesses, synthetic trajectory pipelines, or execution-backed reward environments.\n• High velocity, low ego, and enthusiasm for shipping research improvements directly to millions of active builders within days.\n\nPREFERRED / BONUS EXPERIENCE\n• Specific experience training code-generation LLMs, fast diff/speculative-edit models, or UI-to-code vision-language models.\n• Familiarity with vLLM/SGLang inference optimization, quantization, or online bandit/preference learning from live product traffic.",
      "updated_at": "2026-11-15T17:22:33.074Z",
      "created_at": "2026-11-15T16:55:35.505Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/ai-research-engineer-code-ui-post-training",
        "markdown": "https://jobs.talent247.co/jobs/ai-research-engineer-code-ui-post-training.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/ai-research-engineer-code-ui-post-training/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/ai-research-engineer-code-ui-post-training/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "3e83f524-78ae-4524-b81b-2dc7352e65d3",
      "slug": "senior-multimodal-rag-search-systems-engineer",
      "title": "Senior Multimodal RAG & Search Systems Engineer (Contract)",
      "department": "Applied AI",
      "department_code": "APP",
      "location": "Dublin, Ireland (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Contract",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 95000,
        "max": 125000,
        "currency": "EUR",
        "frequency": "yearly",
        "formatted": "€95k – €125k",
        "subtitle": "annual package"
      },
      "summary": "Senior Multimodal RAG & Search Systems Engineer (Contract) (Applied AI) — Dublin, Ireland · Hybrid · €95k – €125k. THE CLIENT & MANDATE Our client is an…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is an acclaimed European enterprise AI lab building next-generation multimodal knowledge platforms and autonomous workplace agents for global organizations. With R&D roots in Northern Europe and a growing engineering hub in Dublin, their platform connects live across enterprise document repositories, spreadsheets, slide decks, codebases, and recorded meetings so teams can delegate complex analytical and operational workflows to trustworthy agents.\n\nThey have partnered with Talent247 to bring on a Senior Multimodal RAG & Search Systems Engineer in Dublin on an initial 6-month contract (with a strong pathway to permanent conversion or extension). You will join their Retrieval & Knowledge Infrastructure squad to scale real-time, permission-mirrored hybrid search across tens of millions of heterogeneous enterprise artifacts.\n\nKEY RESPONSIBILITIES\n• Architect and scale low-latency hybrid retrieval pipelines combining lexical BM25, dense embeddings, and late-interaction/cross-encoder reranking across text, tables, slide visuals, and meeting transcripts.\n• Build ironclad, zero-leak Access Control List (ACL) security-trimming and permission-graph synchronization across Google Workspace, Microsoft SharePoint, Slack, Jira, and Salesforce connectors.\n• Design incremental ingestion, semantic chunking, layout-aware parsing, and deduplication pipelines capable of reflecting upstream enterprise document edits and permission revocations within seconds.\n• Implement online retrieval telemetry, click/citation feedback loops, and automated LLM-judge evaluation suites to systematically eliminate retrieval blind spots on complex multi-hop queries.\n• Optimize vector database shard placement, quantization settings, caching tiers, and query planning to hold p95 retrieval latency below 80ms at multi-terabyte tenant scale.\n• Collaborate directly with agent-orchestration engineers to expose rich search and filter primitives to autonomous planning loops.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack features Python, Rust, TypeScript, Vespa/Qdrant/Elasticsearch, Kubernetes on GCP, Temporal, and PyTorch embedding services. Operating hybrid in central Dublin, this contract role pays an annualized equivalent of €95,000–€125,000 EUR (structured as competitive daily/contract compensation) with high likelihood of long-term extension or permanent conversion.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of backend software and search/ML systems engineering experience building production search, RAG, or distributed data indexing platforms.\n• Deep hands-on expertise with modern vector and hybrid search engines (Vespa, Qdrant, Weaviate, OpenSearch/Elasticsearch, or pgvector at scale).\n• Strong programming skills in Python and either Rust, Go, or TypeScript for high-concurrency ingestion and query microservices.\n• Direct production experience implementing document-level ACL / RBAC permission filtering in enterprise search or SaaS connector pipelines.\n• Solid grasp of chunking strategies, multimodal PDF/slide parsing, reranking tradeoffs, and quantitative IR evaluation metrics (NDCG, Recall@K, MRR).\n\nPREFERRED / BONUS EXPERIENCE\n• Experience with late-interaction retrieval (ColBERT/ColPali-style visual document retrieval) or custom query-understanding SLMs.\n• Familiarity with Temporal workflows, Kafka/PubSub streaming ingestion, and GCP Kubernetes deployments.",
      "updated_at": "2026-11-15T17:22:23.340Z",
      "created_at": "2026-11-15T16:55:39.096Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-multimodal-rag-search-systems-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-multimodal-rag-search-systems-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-multimodal-rag-search-systems-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-multimodal-rag-search-systems-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7aefccd3-7690-479c-a501-517cd74307d9",
      "slug": "senior-sovereign-llm-inference-engineer-cuda-rocm",
      "title": "Senior Sovereign LLM Inference Engineer (CUDA / ROCm)",
      "department": "AI Infrastructure",
      "department_code": "INF",
      "location": "Montreal, QC, Canada (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 165000,
        "max": 230000,
        "currency": "CAD",
        "frequency": "yearly",
        "formatted": "CA$165k – CA$230k",
        "subtitle": "annual package"
      },
      "summary": "Senior Sovereign LLM Inference Engineer (CUDA / ROCm) (AI Infrastructure) — Montreal, QC, Canada · Hybrid · CA$165k – CA$230k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is Canada's flagship enterprise foundation model laboratory, operating dual research and infrastructure hubs in Montreal and Toronto alongside offices in London and New York. They build frontier multilingual reasoning models, retrieval stacks, and sovereign AI deployments specifically engineered for regulated enterprises, financial institutions, defence organizations, and national governments that require full data custody inside private cloud VPCs or air-gapped on-premise data centers.\n\nThey are working with Talent247 to hire a Senior Sovereign LLM Inference Engineer (CUDA / ROCm) in Montreal. Joining their Inference & Deployment Engineering group, you will own the low-level runtime optimizations and reproducible container packaging that enable massive enterprise LLMs and rerankers to run at peak throughput across heterogeneous NVIDIA and AMD Instinct GPU fleets inside customer-controlled environments.\n\nKEY RESPONSIBILITIES\n• Profile, optimize, and productionize high-throughput LLM serving runtimes (vLLM, TensorRT-LLM, SGLang) across both NVIDIA CUDA (H100/H200/B200) and AMD ROCm (MI300X/MI325X) hardware targets.\n• Implement and tune low-precision quantization schemes (FP8, INT4/AWQ/GPTQ), speculative decoding, chunked prefill, prefix KV-cache reuse, and constrained structured-output decoding.\n• Architect reproducible, hardened, zero-external-dependency Helm and OCI container bundles engineered to deploy cleanly inside air-gapped government and banking Kubernetes clusters.\n• Conduct cross-layer tail-latency investigations from HTTP/gRPC request batching and tokenizer throughput down to GPU kernel launch overhead and tensor/pipeline parallelism communication.\n• Build automated hardware-matrix benchmark harnesses measuring TTFT, ITL, tokens-per-GPU-hour, and accuracy retention across every model release.\n• Partner with forward-deployed architects and enterprise infrastructure teams to size GPU footprints and resolve high-scale production bottlenecks.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack spans C++17/20, Python, PyTorch, CUDA, HIP/ROCm, Triton, vLLM, TensorRT-LLM, Docker, and Kubernetes. Offered on a hybrid basis from Montreal, compensation includes CA$165,000–CA$230,000 base salary, competitive equity options, comprehensive Quebec/Canadian benefits, and full visa/relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of systems, HPC, or machine learning infrastructure engineering focused on GPU acceleration and production model serving.\n• Strong proficiency in C++ and Python, paired with hands-on experience profiling and optimizing CUDA, Triton, or AMD ROCm/HIP workloads.\n• Direct production experience operating or contributing to modern LLM inference engines such as vLLM, TensorRT-LLM, SGLang, or Triton Inference Server.\n• Deep understanding of transformer memory bandwidth bottlenecks, KV-cache management, quantization tradeoffs, and multi-GPU tensor parallelism.\n• Solid command of Linux systems internals, Docker/OCI containers, and Kubernetes GPU operator deployments in security-sensitive environments.\n\nPREFERRED / BONUS EXPERIENCE\n• Experience deploying enterprise software into air-gapped, VPC-isolated, or government/financial-services environments.\n• Upstream open-source contributions to vLLM, SGLang, PyTorch, FlashAttention, or ROCm libraries.",
      "updated_at": "2026-11-15T17:22:01.792Z",
      "created_at": "2026-11-15T16:55:45.970Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-sovereign-llm-inference-engineer-cuda-rocm",
        "markdown": "https://jobs.talent247.co/jobs/senior-sovereign-llm-inference-engineer-cuda-rocm.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-sovereign-llm-inference-engineer-cuda-rocm/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-sovereign-llm-inference-engineer-cuda-rocm/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "7ed5c2e0-79e4-4cbe-af58-33c0c0546a4f",
      "slug": "senior-research-scientist-generative-sensor",
      "title": "Senior Research Scientist, Generative Sensor Simulation & Physical AI",
      "department": "Robotics & Embodied AI",
      "department_code": "ROB",
      "location": "Toronto, ON, Canada (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 185000,
        "max": 265000,
        "currency": "CAD",
        "frequency": "yearly",
        "formatted": "CA$185k – CA$265k",
        "subtitle": "annual package"
      },
      "summary": "Senior Research Scientist, Generative Sensor Simulation & Physical AI (Robotics & Embodied AI) — Toronto, ON, Canada · Hybrid · CA$185k – CA$265k. THE…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a globally recognized physical-AI pioneer headquartered in Toronto, backed by premier technology and logistics investors to bring AI-first autonomous commercial trucking to public highways. Their core competitive moat is a generative neural simulator capable of constructing photorealistic, physically consistent digital twins of real-world highways, complete with reactive human-like traffic agents and multi-modal LiDAR/camera sensor synthesis.\n\nThey have engaged Talent247 to recruit a Senior Research Scientist, Generative Sensor Simulation & Physical AI in Toronto. Embedded within their multidisciplinary Research & Autonomy team, you will invent and scale next-generation 3D neural rendering, generative video/LiDAR world models, and adversarial traffic-agent policies that teach their autonomous driver to handle safety-critical highway scenarios before a truck ever encounters them on the road.\n\nKEY RESPONSIBILITIES\n• Formulate and execute an ambitious research agenda across 3D neural scene reconstruction (3D Gaussian Splatting, NeRFs, neural meshes), multi-sensor physics synthesis, and spatiotemporal diffusion/flow-matching world models.\n• Develop closed-loop, language- and goal-conditioned traffic agent models capable of generating realistic, interactive, and adversarial safety-critical maneuvers around the ego vehicle.\n• Address cross-sensor consistency across synchronized multi-camera rigs, long-range LiDAR point clouds, and radar returns under dynamic lighting and adverse weather conditions.\n• Train and scale large generative vision and trajectory models across multi-node GPU clusters, partnering closely with research engineers to transition breakthroughs into nightly simulation validation suites.\n• Quantify and close the sim-to-real domain gap through rigorous perceptual, geometric, and downstream closed-loop policy correlation metrics.\n• Publish seminal work at premier venues where appropriate while mentoring junior scientists and shaping technical direction across physical AI.\n\nTECHNICAL ENVIRONMENT & OFFERING\nResearchers work in a unified PyTorch, C++, and CUDA codebase backed by dedicated multi-node H100 clusters and petabyte-scale highway sensor logs. Working hybrid in downtown Toronto, our client provides CA$185,000–CA$265,000 base compensation plus substantial equity participation, comprehensive health benefits, and Canadian visa/relocation sponsorship.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• PhD in Computer Vision, Machine Learning, Robotics, Computer Graphics, or equivalent frontier industrial research track record.\n• First-author publications at top-tier conferences such as CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, SIGGRAPH, RSS, or CoRL.\n• Deep hands-on expertise in 3D neural rendering (Gaussian Splatting, NeRFs, implicit surfaces), video/LiDAR diffusion models, or multi-agent trajectory modeling in PyTorch.\n• Strong mathematical foundations in multi-view 3D geometry, differentiable rendering, sensor physics, or reinforcement/imitation learning.\n• Ability to write clean, performant research code in Python (and C++/CUDA where needed) and validate hypotheses at distributed compute scale.\n\nPREFERRED / BONUS EXPERIENCE\n• Direct industry experience with autonomous driving perception, closed-loop neural simulation, or generative world models.\n• Experience optimizing real-time rendering or generative inference kernels in CUDA, OptiX, or TensorRT.",
      "updated_at": "2026-11-15T17:21:52.337Z",
      "created_at": "2026-11-15T16:55:48.732Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-research-scientist-generative-sensor",
        "markdown": "https://jobs.talent247.co/jobs/senior-research-scientist-generative-sensor.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-generative-sensor/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-research-scientist-generative-sensor/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "b8244fcf-f5ec-4772-9aed-58d3ec574961",
      "slug": "senior-simulation-compute-mlops-platform-engineer",
      "title": "Senior Simulation Compute & MLOps Platform Engineer",
      "department": "MLOps & Platform",
      "department_code": "OPS",
      "location": "Toronto, ON, Canada (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 160000,
        "max": 215000,
        "currency": "CAD",
        "frequency": "yearly",
        "formatted": "CA$160k – CA$215k",
        "subtitle": "annual package"
      },
      "summary": "Senior Simulation Compute & MLOps Platform Engineer (MLOps & Platform) — Toronto, ON, Canada · Hybrid · CA$160k – CA$215k. THE CLIENT & MANDATE Our…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a physical-AI unicorn headquartered in downtown Toronto developing next-generation autonomous freight technology built from the ground up around a closed-loop generative neural simulator. Rather than logging tens of millions of brute-force highway miles to discover safety-critical edge cases, their platform synthesizes reactive traffic worlds and sensor-accurate scenarios in software to train and stress-test end-to-end driving models continuously.\n\nThey are partnering with Talent247 to hire a Senior Simulation Compute & MLOps Platform Engineer in Toronto. Joining their Simulation & ML Infrastructure group, you will own the high-throughput GPU orchestration engine that executes millions of closed-loop virtual miles overnight, ensuring research scientists and autonomy engineers receive deterministic, reproducible safety signals on every model commit.\n\nKEY RESPONSIBILITIES\n• Architect, scale, and operate distributed Kubernetes and Ray GPU compute pipelines executing tens of thousands of concurrent neural sensor-simulation and closed-loop evaluation jobs daily.\n• Design deterministic simulation replay, artifact caching, and frame-level debugging tooling so autonomy engineers can reproduce exact multi-agent highway edge cases in seconds.\n• Optimize GPU cluster scheduling, bin-packing, container cold-start times, and NVMe/object-store streaming throughput across cloud and dedicated accelerator footprints.\n• Build automated release-qualification gates, safety telemetry dashboards, and regression-bisect workflows linking model checkpoints directly to closed-loop driving KPIs.\n• Partner with neural rendering and driving-model researchers to profile CUDA/PyTorch bottlenecks across coupled simulator-and-driver execution graphs.\n• Drive infrastructure reliability, Infrastructure-as-Code discipline, cost-per-simulated-mile telemetry, and on-call automation across the ML platform estate.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir engineering environment combines Python, Rust, C++, PyTorch, Kubernetes, Ray, Argo/Flyte, Prometheus/OpenTelemetry, and multi-cloud GPU clusters. Based hybrid in Toronto, our client offers CA$160,000–CA$215,000 base salary alongside competitive equity, full Canadian medical/dental/vision benefits, and relocation support.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of experience in MLOps, ML platform engineering, distributed compute orchestration, or large-scale simulation infrastructure.\n• Production mastery of Kubernetes at scale, including custom operators/controllers, GPU scheduling, container runtime optimization, and multi-node job orchestration.\n• Strong software engineering proficiency in Python alongside Rust, Go, or C++ for performance-critical control-plane and data-plane services.\n• Proven experience building high-throughput batch/streaming ML evaluation pipelines, artifact registries, and observability stacks (Prometheus, Grafana, OpenTelemetry).\n• Track record diagnosing cross-layer performance bottlenecks across GPU memory, network I/O, object storage, and Linux host configuration.\n\nPREFERRED / BONUS EXPERIENCE\n• Background supporting autonomous vehicles, robotics simulation, game-engine/neural-rendering farms, or HPC Slurm/Ray clusters.\n• Familiarity with PyTorch distributed execution, TensorRT, CUDA profiling tools (Nsight Systems), and CI/CD hardware-in-the-loop or sil-in-the-loop testing.",
      "updated_at": "2026-11-15T17:21:40.805Z",
      "created_at": "2026-11-15T16:55:49.380Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-simulation-compute-mlops-platform-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-simulation-compute-mlops-platform-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-simulation-compute-mlops-platform-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-simulation-compute-mlops-platform-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    },
    {
      "id": "b618a080-b15d-4bbc-96ff-9790e4c0515a",
      "slug": "senior-nlp-scientific-extraction-ml-engineer",
      "title": "Senior NLP & Scientific Extraction ML Engineer",
      "department": "Machine Learning",
      "department_code": "ML",
      "location": "Montreal, QC, Canada (Hybrid)",
      "workplace_type": "Hybrid",
      "employment_type": "Full-time",
      "is_remote": false,
      "is_featured": false,
      "headcount": 1,
      "salary": {
        "min": 155000,
        "max": 210000,
        "currency": "CAD",
        "frequency": "yearly",
        "formatted": "CA$155k – CA$210k",
        "subtitle": "annual package"
      },
      "summary": "Senior NLP & Scientific Extraction ML Engineer (Machine Learning) — Montreal, QC, Canada · Hybrid · CA$155k – CA$210k. THE CLIENT & MANDATE Our client is…",
      "description_markdown": "THE CLIENT & MANDATE\nOur client is a well-funded applied artificial intelligence company rooted in Montreal's premier deep-learning research ecosystem, with an expanding European engineering presence. They build specialized, zero-hallucination tabular extraction and scientific literature synthesis models trusted by global biopharmaceutical clinical, regulatory, and market-access teams, where a single misread clinical endpoint or omitted cohort statistic is unacceptable.\n\nThey have retained Talent247 to identify a Senior NLP & Scientific Extraction ML Engineer for their core modeling group in Montreal. Sitting alongside research scientists and production platform engineers, this hire owns the domain-tuned small language models (SLMs), multimodal document parsers, and neuro-symbolic citation verifiers that turn thousands of complex PDF clinical trial studies, regulatory filings, and nested statistical tables into provenance-linked structured datasets.\n\nKEY RESPONSIBILITIES\n• Architect, post-train, and fine-tune specialized language and vision-language models (1B–32B parameters) using SFT, DPO, and verifiable-reward RL to extract structured clinical and scientific entities from dense PDFs and complex multi-page tables.\n• Design and maintain deterministic constrained-decoding pipelines and neuro-symbolic verification layers that bind every extracted figure, confidence interval, and claim back to exact character spans and bounding-box coordinates in source documents.\n• Build high-recall, multi-stage biomedical retrieval and cross-encoder reranking pipelines capable of screening millions of peer-reviewed publications and regulatory submissions without false negatives.\n• Create rigorous, contamination-free evaluation harnesses partnered with domain biostatisticians and medical reviewers to measure cell-level extraction precision, recall, and hallucination rates prior to production promotion.\n• Collaborate closely with inference and backend engineers to distil, quantize, and serve custom models within strict enterprise latency and GPU cost budgets.\n• Diagnose systematic failure modes across OCR, layout segmentation, and long-context reasoning, turning real-world customer edge cases into high-signal synthetic and human-verified training curricula.\n\nTECHNICAL ENVIRONMENT & OFFERING\nTheir stack centers on Python, PyTorch, custom VLM/layout transformers, vLLM/TensorRT-LLM serving, Ray Data, and Kubernetes-backed H100 clusters. Operating on a collaborative hybrid schedule out of downtown Montreal, our client offers CA$155,000–CA$210,000 base salary plus meaningful early-stage equity, comprehensive Canadian health and dental coverage, conference publication support, and relocation/visa sponsorship for exceptional talent.",
      "requirements_markdown": "ESSENTIAL QUALIFICATIONS\n• 5+ years of hands-on machine learning and NLP engineering experience shipping custom-trained transformer models into production environments.\n• Deep fluency in Python and PyTorch, including custom training loops, mixed-precision distributed training (FSDP/DeepSpeed), LoRA/PEFT, and preference or reinforcement post-training.\n• Proven track record in structured information extraction, document layout understanding, table-parsing vision-language models, or grammar-constrained generation.\n• Strong command of retrieval-augmented architectures, dense/sparse biomedical embeddings, neural reranking, and high-recall evaluation methodology (precision/recall calibration, abstention modeling).\n• Disciplined software engineering habits across version-controlled ML pipelines, automated regression testing, and reproducible dataset lineage.\n\nPREFERRED / BONUS EXPERIENCE\n• Prior exposure to biomedical, clinical-trial, pharmacovigilance, or scientific literature corpora (PubMed, ClinicalTrials, FDA/EMA submissions).\n• Hands-on experience with inference optimization, model distillation, or FP8/INT4 quantization using vLLM, SGLang, or TensorRT-LLM.\n• MSc or PhD in Machine Learning, Computational Linguistics, Bioinformatics, or Computer Science, or publications at ACL, EMNLP, NAACL, NeurIPS, or ICML.",
      "updated_at": "2026-11-15T17:21:31.294Z",
      "created_at": "2026-11-15T16:55:49.880Z",
      "urls": {
        "html": "https://jobs.talent247.co/jobs/senior-nlp-scientific-extraction-ml-engineer",
        "markdown": "https://jobs.talent247.co/jobs/senior-nlp-scientific-extraction-ml-engineer.md",
        "form_schema": "https://jobs.talent247.co/api/public/jobs/senior-nlp-scientific-extraction-ml-engineer/form-schema",
        "apply": "https://jobs.talent247.co/api/public/jobs/senior-nlp-scientific-extraction-ml-engineer/apply"
      },
      "webmcp": {
        "declarative_tool": "apply_for_role",
        "required_fields": [
          "full_name",
          "email",
          "consent"
        ],
        "cv_requirements": {
          "allowed_extensions": [
            ".pdf",
            ".docx"
          ],
          "max_bytes": 10485760,
          "alternative_field": "linkedin_url"
        }
      }
    }
  ]
}