Senior ML Engineer — LLM Inference Optimization

FDE Times; Greater London; Hybrid; GBP 90,000 - 150,000; Full time

Benefits offered by this job

Competitive compensation

Career growth

Flexibility and ownership

Job summary

Nebius is seeking a Senior Machine Learning Engineer to own model and endpoint optimization from artifacts to production deployment on the Applied AI team. You will tackle quantization, model compression, speculative decoding, KV-cache optimization, and multi-node GPU execution to boost latency, throughput, and cost per token while preserving quality.

You will collaborate with kernel and platform engineers to investigate serving problems, compare configurations, and validate improvements through

Qualifications

• Strong Python and PyTorch skills.
• Hands-on experience deploying or optimizing LLM/VLM or high-throughput transformer inference.
• Experience with at least one modern inference stack (vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve or KServe).
• Understanding of transformer inference bottlenecks including KV cache, memory bandwidth, and long-context serving.

Responsibilities

• Own optimization projects for specific model families, customer endpoints, or serving backends.
• Compare inference engines and recommend configurations for workloads.
• Diagnose model-quality and performance regressions during production rollouts.
• Improve LLM and VLM endpoint latency, throughput, memory efficiency, and cost per token.
• Deploy, benchmark, and extend inference engines and deployment stacks.
• Design and implement efficient KV-cache and prefill/decode strategies.
• Collaborate with kernel and platform engineers to trace bottlenecks across code paths.

Skills

Python

PyTorch

LLM inference

VLM inference

Distributed inference

Communication

Tools

vLLM

SGLang

TensorRT-LLM

Triton Inference Server

NVIDIA Dynamo

Ray Serve

KServe

Job description

Nebius is seeking a Senior Machine Learning Engineer to own model and endpoint optimization from artifacts to production deployment on the Applied AI team. You will tackle quantization, model compression, speculative decoding, KV-cache optimization, and multi-node GPU execution to boost latency, throughput, and cost per token while preserving quality.

You will collaborate with kernel and platform engineers to investigate serving problems, compare configurations, and validate improvements through