Senior ML Engineer, LLM Inference Optimization

Lever, Inc.; Netherlands; On-site; EUR 120,000 - 180,000; Full time

Benefits offered by this job

Competitive compensation

Career growth opportunities

Flexibility and ownership over work

Collaborative environment

Exposure to advanced AI infrastructure

Job summary

Lever, Inc. is seeking a Senior Machine Learning Engineer focused on LLM/VLM inference optimization based in the Netherlands. You will drive latency, throughput, memory efficiency, and cost-per-token improvements across modern inference stacks.

The role demands hands-on optimization, benchmarking, and collaboration with kernel, platform, and research teams to push production readiness. The candidate should excel in Python and PyTorch, with experience deploying or optimizing high-throughput

Qualifications

• Strong Python and PyTorch software skills.
• Hands-on experience with deploying or optimizing high-throughput transformer models.
• Familiarity with modern inference stacks (e.g., vLLM, Triton, TensorRT-LLM).

Responsibilities

• Own optimization initiatives for LLM/VLM inference across model families and endpoints.
• Evaluate serving configurations to meet latency and throughput targets.
• Diagnose production performance regressions and implement improvements.
• Develop reproducible benchmarks for tokens per second, latency percentiles, and memory usage.
• Collaborate with kernel, platform, and research teams to drive deployments.

Skills

Python

PyTorch

LLM Inference

Performance optimization

Benchmarking

Tools

vLLM

SGLang

TensorRT-LLM

Triton Inference Server

Ray Serve

KServe

CUDA

Job description

Lever, Inc. is seeking a Senior Machine Learning Engineer focused on LLM/VLM inference optimization based in the Netherlands. You will drive latency, throughput, memory efficiency, and cost-per-token improvements across modern inference stacks.

The role demands hands-on optimization, benchmarking, and collaboration with kernel, platform, and research teams to push production readiness. The candidate should excel in Python and PyTorch, with experience deploying or optimizing high-throughput