Senior Machine Learning Engineer, LLM Inference Optimization

Fdetimes; Amsterdam; Híbrido; EUR 120.000 - 180.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Competitive compensation

Career growth and learning

Flexibility and ownership

Collaborative and innovative culture

Impactful AI projects

International environment and talented

Descripción de la vacante

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 model internals, quantization, speculative decoding, KV-cache optimization, and multi-node GPU strategies to reduce latency, boost throughput, and lower token成本 while preserving quality.

You will collaborate with kernel and platform engineers to diagnose performance regressions, evaluate engine configurations, and validate

Formación

• Strong engineering skills in Python and PyTorch.
• Hands-on experience deploying or optimizing LLM/VLM or high-throughput transformer inference.
• Knowledge of modern inference stacks (vLLM, Triton, TensorRT-LLM, SGLang, NVIDIA Dynamo, Ray Serve or KServe).
• Understanding transformer inference bottlenecks: KV cache, memory bandwidth, batching, and long-context serving.
• Ability to quantify latency/throughput trade-offs and cost.
• Experience designing distributed inference system architectures (routing, scheduling, PDD, multi-node inference, or expert parallelism).
• Strong communication and collaboration across research, kernel, infrastructure, product, and customers.

Responsabilidades

• Own optimization projects for specific model families, endpoints, or serving backends.
• Compare inference engines and configure best-serving setups per workload.
• Diagnose model-quality and performance regressions during production rollouts.
• Improve LLM/VLM endpoint latency, throughput, memory efficiency, GPU utilization, and cost per token.
• Deploy, benchmark, and extend engines (e.g., vLLM, Triton, TensorRT-LLM, Triton Server, CUDA) in production.
• Develop model-compression workflows including quantization and distillation.
• Implement speculative decoding, KV-cache optimization, and prefill/decode strategies.
• Design and tune LLM request routers and scheduling to balance latency and reliability.
• Scale dense and Mixture-of-Experts inference across multiple GPU nodes with proper parallelism.

Conocimientos

Python

PyTorch

LLM/Transformer inference

Distributed inference architecture

Inference stacks (vLLM/Triton/Ray)

Herramientas

vLLM

SGLang

TensorRT-LLM

Triton Inference Server

NVIDIA Dynamo

Ray Serve

KServe

Descripción del empleo

About Nebius

Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure.

Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI.

Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D.

The role

As a Senior Machine Learning Engineer on the Applied AI team at Nebius Token Factory, you will own model and endpoint optimization from model artifacts through production deployment. Your work will span model internals, quantization and model compression, speculative decoding, KV-cache optimization, inference engines, serving architecture, and benchmarking to improve latency, throughput, memory efficiency, GPU utilization, and cost per token while preserving quality and reliability.

You will work with kernel and platform engineers to investigate serving problems, compare configurations, and resolve performance and quality regressions. You will evaluate model- and engine-level optimizations together with distributed inference designs, including request routing, scheduling, prefill/decode coordination, and multi-node GPU execution. Your improvements will be validated through reproducible benchmarks and safe production rollouts under real-world workloads.

Your responsibilities

• Own optimization projects for specific model families, customer endpoints, or serving backends.
• Compare inference engines and recommend serving configurations suited to individual workloads.
• Diagnose model-quality and performance regressions during production rollouts.
• Improve LLM and VLM endpoint latency, throughput, memory efficiency, GPU utilization, quality, and cost per token.
• Deploy, configure, benchmark, and extend engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, and NVIDIA Dynamo.
• Develop production model-compression workflows covering quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery.
• Implement or integrate speculative decoding, draft-model approaches, KV-cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. For prefill–decode disaggregation (PDD), evaluate KV-cache transfer, worker placement, and capacity balancing to determine which workloads benefit from the architecture.
• Design and improve LLM request routers and scheduling policies that balance worker utilization, queueing, request characteristics, and KV-cache locality while meeting latency and reliability targets.
• Scale dense and mixture-of-experts inference across multiple GPU nodes. Select and tune tensor, pipeline, data, and expert parallelism—including wide expert parallelism (WideEP)—with attention to hardware topology, expert load balance, and communication overhead.
• Build reproducible benchmark harnesses measuring TTFT, TPOT, tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Validate architecture choices under representative traffic and consistent GPU budgets.
• Collaborate with GPU kernel and platform engineers to trace bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers.
• Produce design documents, performance reports, rollout plans, and customer-facing technical explanations.

Must-haves

• Strong engineering skills in Python and PyTorch.
• Hands‑on experience deploying or optimizing LLM, VLM, or high‑throughput transformer inference.
• Practical knowledge of at least one modern inference stack, such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or an equivalent internal system.
• A strong understanding of transformer inference bottlenecks involving KV cache, attention, memory bandwidth, batching, parallelism, and long‑context serving.
• Ability to quantify trade‑offs among latency, throughput, quality, utilization, and cost.
• Experience designing or optimizing distributed inference system architecture, with hands‑on work in one or more areas such as request routing, distributed scheduling, PDD, multi‑node inference, or expert parallelism.
• Strong communication and collaboration skills across research, kernel, infrastructure, product, and customer teams.

Nice‑to‑haves

• Experience with quantization‑aware training, post‑training quantization, FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, or SmoothQuant.
• Experience with distillation, speculative decoding, EAGLE, Medusa, multi‑token prediction, or related inference acceleration methods.
• Experience supporting agentic workloads involving tool calling, structured outputs, streaming APIs, high concurrency, and multi‑step orchestration.
• Familiarity with CUDA or Triton; the role does not require kernel engineering to be the candidate’s primary specialization.
• Contributions to open‑source projects such as vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, or KServe.
• Implementation experience with cache‑aware request routing, PDD, or WideEP, including diagnosing communication bottlenecks and load imbalance.
• Familiarity with GPU communication libraries and interconnects, such as NCCL, NVLink, InfiniBand, or RoCE.

Benefits & Perks

• Competitive compensation
• Career growth and learning opportunities
• Flexibility and ownership
• Collaborative and innovative culture
• Opportunity to work on impactful AI projects
• International environment and talented teams

What's it like to work at Nebius

Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI

Equal Opportunity Statement

Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law.

Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire.

If you need accommodations during the application process, please let us know.