Lead AI Inference Performance Engineer

Showcify, Inc.; United States; Presencial; USD 140 000 - 200 000; Tempo integral

Resumo da oferta

d-Matrix is seeking a Senior Staff Inference Performance Engineer to own end-to-end performance of deployed workloads in an internal cloud and customer-facing inference endpoints. You will profile, hypothesize, test, and hand off fixes with evidence to owning teams, ensuring performance parity with bare metal across heterogeneous architectures.

You will diagnose across orchestration, networking, and inference internals, isolate layers, and craft minimal problem statements for cross-stack teams

Qualificações

• 8+ years of hands-on performance or systems engineering.
• Track record of finding performance overhead through profiling in unfamiliar, sparsely documented systems.
• Deep systems fundamentals including Linux internals, NUMA, CPU pinning and cpusets, PCIe, memory and cache behavior, and network paths.
• Experience running performance-sensitive workloads on Kubernetes and identifying latency/throughput costs.
• Working knowledge of LLM inference serving, batching and KV-cache.
• Proficiency in Python and C/C++, to instrument, debug and patch code across the stack.
• Clear technical writing that hardware/firmware teams can act on without extra context.

Responsabilidades

• Diagnose performance problems across the full serving path from orchestration to inference internals.
• Isolate problems to specific layers and determine fix routes or handoffs.
• Write minimal, reproducible problem statements for cross-stack findings.
• Protect throughput, latency, and stability targets against bare metal baselines.
• Assess scalability from single deployments to fleet operation, including routing and shared-resource contention.
• Capture workload requirements in the orchestration layer configuration to ensure performant deployment by default.
• Build lightweight telemetry and benchmarks for prototype stacks.
• Turn recurring investigations into playbooks and mentor engineers in profiling-first methods.

Conhecimentos

Performance engineering

Linux internals

NUMA

Kubernetes

Python

C/C++

Profiling

TensorRT

Ferramentas

CUDA

Nsight

NCCL

vLLM

Descrição da oferta de emprego

d-Matrix is seeking a Senior Staff Inference Performance Engineer to own end-to-end performance of deployed workloads in an internal cloud and customer-facing inference endpoints. You will profile, hypothesize, test, and hand off fixes with evidence to owning teams, ensuring performance parity with bare metal across heterogeneous architectures.

You will diagnose across orchestration, networking, and inference internals, isolate layers, and craft minimal problem statements for cross-stack teams