---
title: "Senior MLOps & Enterprise LLM Serving Platform Engineer"
slug: "senior-mlops-enterprise-llm-serving-platform"
id: "06be68f5-2a3c-4bec-9c6f-0c2133b0e077"
department: "MLOps & Platform"
department_code: "OPS"
location: "Berlin / Cologne, Germany"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "€92k – €125k (annual package)"
is_remote: false
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform"
markdown_url: "https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:24:09.782Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Senior MLOps & Enterprise LLM Serving Platform Engineer

## Position Telemetry

- **Reference:** `SENIOR-MLOPS-ENTER`
- **Practice / Sector:** MLOps & Platform (`OPS`)
- **Location:** Berlin / Cologne, Germany
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** €92k – €125k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform

## Role Dossier & Specification

THE CLIENT & MANDATE
Our 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.

They 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.

KEY RESPONSIBILITIES
• 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.
• Build zero-downtime model rollout workflows featuring shadow traffic mirroring, automated quality/latency canary gates, instant rollback, and cryptographic artifact provenance.
• 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.
• Instrument deep GPU and inference observability (TTFT, TPOT, KV-cache pressure, SM occupancy, NVLink throughput) using Prometheus, OpenTelemetry, DCGM, and Grafana.
• Partner directly with research scientists and kernel engineers to streamline the path from multi-node training checkpoint to quantized production endpoint.
• Drive GitOps automation, disaster-recovery readiness, and strict zero-retention enterprise compliance across European and North American serving regions.

TECHNICAL ENVIRONMENT & OFFERING
Their 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.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 5+ years of MLOps, ML platform, or production infrastructure engineering experience operating high-throughput distributed services on Kubernetes.
• Direct production experience serving large transformer models using vLLM, TensorRT-LLM, Triton Inference Server, SGLang, or custom C++/CUDA runtimes.
• Strong programming skills in Python and Go (or Rust) for writing Kubernetes operators, deployment controllers, and performance benchmarking tools.
• Deep operational familiarity with Linux networking, NVIDIA GPU Operator/MIG/DCGM telemetry, Envoy/gRPC load balancing, and GitOps (ArgoCD/Flux, Terraform).
• Proven ability to diagnose complex tail-latency and throughput regressions across model serving stacks.

PREFERRED / BONUS EXPERIENCE
• Experience operating bare-metal HPC/Slurm or multi-thousand-GPU data center environments alongside cloud Kubernetes clusters.
• Working knowledge of model quantization (FP8/INT4), speculative decoding, or disaggregated prefill/decode serving architectures.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/senior-mlops-enterprise-llm-serving-platform or via HTTP `POST https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/senior-mlops-enterprise-llm-serving-platform/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
