Staff Machine Learning Engineer, LLM Inference Optimization

pbAbout Nebius: /b /ppNebius 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. /ppBuilt by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. /ppListed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with RD 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 RD. /ph3Role Summary /h3pNebius Token Factory is building production-grade inference for open-source, customer, and internally optimized models. This role owns the path from model artifact to fast, reliable, cost-efficient endpoint. The work sits at the intersection of model internals, inference engines, serving architecture, benchmarking, and production deployment. /ph3What You Will Do /h3ullipOwn a platform-level optimization pillar such as quantization, speculative decoding, KV-cache optimization, engine selection, or long-context serving. /p /lilipCreate reusable evaluation, benchmarking, and rollout frameworks across multiple model families. /p /lilipProvide technical direction to MLEs, Applied Scientists, GPU kernel engineers, and infrastructure engineers on shared performance goals. /p /lilipOptimize LLM and VLM endpoints for latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. /p /lilipDeploy, configure, benchmark, and extend inference engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, or similar systems. /p /lilipBuild and productionize model-compression workflows, including quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. /p /lilipImplement or integrate speculative decoding, draft-model approaches, KV-cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. /p /lilipBuild reproducible benchmark harnesses for TTFT, TPOT, tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. /p /lilipPartner with GPU kernel engineers and platform engineers to diagnose bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. /p /lilipWrite clear design docs, performance reports, rollout plans, and customer-facing technical explanations. /p /li /ulh3Required Qualifications /h3ullipStrong Python and PyTorch engineering skills. /p /lilipHands-on experience deploying or optimizing LLM, VLM, or high-throughput transformer inference systems. /p /lilipPractical knowledge of at least one modern inference stack such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or equivalent internal systems. /p /lilipStrong understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. /p /lilipAbility to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. /p /lilipStrong communication skills and ability to collaborate with research, kernel, infrastructure, product, and customer teams. /p /li /ulh3Preferred Qualifications /h3ullipExperience with quantization-aware training, post-training quantization, FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, SmoothQuant, or related techniques. /p /lilipExperience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods. /p /lilipExperience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. /p /lilipCUDA or Triton familiarity, even if the role is not primarily a kernel-engineering role. /p /lilipOpen-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects. /p /li /ulpbBenefits Perks: /b /pulliCompetitive compensation /liliCareer growth and learning opportunities /liliFlexibility and ownership /liliCollaborative and innovative culture /liliOpportunity to work on impactful AI projects /liliInternational environment and talented teams /li /ulpbWhat's it like to work at Nebius: /b /ppFast moving- Bold thinking- Constant growth- Meaningful impact- Trust and real ownership- Opportunity to shape the future of AI /ppbEqual Opportunity Statement: /b /ppNebius 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. /ppApplicants 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. /ppIf you need accommodations during the application process, please let us know. /ppFind Jobs in Switzerland on Arbeitnow /p #J-18808-Ljbffr