Skip to main content
TALENT247 // SPECIALIST AI EXECUTIVE SEARCH & TECHNICAL RECRUITMENT
INF · AI Infrastructure

Senior C++ Engine Engineer, AI-Native Query Execution

Seattle, WA · Full-time · Hybrid

AI Specialism

AI Infrastructure

Workplace

Hybrid

Location

Seattle, WA

Compensation

$210k – $295k

annual package

Role Dossier & Specification

REF: SENIOR-C-ENGINE-EN

THE CLIENT & MANDATE

Our 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.

They 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.

KEY RESPONSIBILITIES

  • 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.
  • 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.
  • 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.
  • 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.
  • 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.
  • Profile CPU cache misses, memory allocator contention, and async coroutine scheduling across thousand-node analytical clusters.

TECHNICAL ENVIRONMENT & OFFERING

Stack: 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.

Core Requirements

ESSENTIAL QUALIFICATIONS

  • 5+ years of systems engineering experience in modern C++ (C++17/20) building database execution engines, compilers, storage engines, or HPC frameworks.
  • Deep understanding of vectorized columnar execution, Volcano vs. push-based pipelines, cache-conscious data structures, SIMD intrinsics, and custom memory management.
  • Production familiarity with Apache Arrow, Parquet, ORC, or similar columnar memory/file formats and query optimization principles.
  • Strong concurrency and asynchronous systems skills (lock-free queues, C++20 coroutines/futures, high-throughput RPC clients).
  • Demonstrated rigor in micro-benchmarking, hardware performance counter analysis, and deterministic fuzz/correctness testing.

PREFERRED / BONUS EXPERIENCE

  • Experience integrating vector search (FAISS, HNSW, DiskANN) or batch LLM inference scheduling directly into analytical query engines (Velox, DuckDB, DataFusion, ClickHouse).
  • Knowledge of LLM serving mechanics (continuous batching, prefix caching, structured JSON decoding) and how query ordering impacts cache hit rates.
  • Contributions to major open-source database or columnar data projects.

Position Telemetry

Location Seattle, WA (Hybrid)
Department AI Infrastructure
Employment type Full-time
Workplace type Hybrid
Salary range
$210k – $295k annual package