Staff ML Engineer - Inference Platform & Optimization

Fdetimes; Amsterdam; Remote; EUR 120.000 - 180.000; Vollzeit

Benefits dieser Stelle

Competitive compensation

Career growth and learning

Flexibility and ownership

Collaborative culture

Work on impactful AI projects

International environment

Zusammenfassung

Nebius is building production-grade inference for open-source, customer, and internally optimized models. You will own the path from model artifact to fast, reliable, cost-efficient endpoints at platform level, interfacing with MLEs, GPU kernel engineers, and infrastructure teams.

You will optimize end-to-end latency, throughput, and memory usage, deploy engines such as vLLM, TensorRT-LLM, or Triton, and drive reproducible benchmarks across multiple model families.

Qualifikationen

• Strong Python and PyTorch engineering skills.
• Hands-on experience deploying or optimizing LLM/VLM inference systems.

Aufgaben

• Own platform-level optimization pillars such as quantization or long-context serving.
• Create reusable evaluation, benchmarking, and rollout frameworks for multiple model families.
• Provide technical direction to engineers on performance goals.
• Optimize LLM and VLM endpoints for latency, throughput, memory, GPU usage, cost per token.
• Deploy and benchmark inference engines (e.g., vLLM, Triton).

Kenntnisse

Python

PyTorch

LLM deployment

Performance tuning

Tools

vLLM

SGLang

TensorRT-LLM

Triton Inference Server

NVIDIA Dynamo

Ray Serve

KServe

Jobbeschreibung

Nebius is building production-grade inference for open-source, customer, and internally optimized models. You will own the path from model artifact to fast, reliable, cost-efficient endpoints at platform level, interfacing with MLEs, GPU kernel engineers, and infrastructure teams.

You will optimize end-to-end latency, throughput, and memory usage, deploy engines such as vLLM, TensorRT-LLM, or Triton, and drive reproducible benchmarks across multiple model families.