Роль четко определена с акцентом на передовые технологии ИИ, но отсутствие информации о зарплате затрудняет полную оценку привлекательности.
Сильный стек технологий
Продуктовая компания
Нет информации о зарплате
Возможности карьерного роста
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior Machine Learning Engineer (LLM Inference Optimization): Optimizing LLM and VLM models and production endpoints with an accent on quantization, inference engines, distributed serving, and GPU efficiency. Focus on designing request routing and scheduling, scaling inference across multi-node GPU systems, building reproducible benchmarks, and resolving latency, quality, reliability, and cost trade-offs.
Location: Zurich, Switzerland
Company
Nebius builds a full-stack AI cloud platform for model training, inference, and production deployment, with engineering hubs across Europe, the UK, North America, and Israel.
What you will do
Own optimization projects for model families, customer endpoints, and inference backends.
Deploy, configure, benchmark, and extend inference engines including vLLM, SGLang, TensorRT-LLM, Triton Inference Server, and NVIDIA Dynamo.
Improve LLM and VLM latency, throughput, memory efficiency, GPU utilization, quality, reliability, and cost per token.
Design distributed inference architectures, including request routing, scheduling, prefill/decode coordination, KV-cache optimization, and multi-node GPU execution.
Build reproducible benchmark harnesses and collaborate with kernel and platform engineers to diagnose bottlenecks across model, runtime, scheduler, gateway, and cluster layers.
Requirements
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, or KServe.
Strong understanding of transformer inference bottlenecks involving KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving.
Experience designing or optimizing distributed inference systems, including request routing, scheduling, prefill/decode disaggregation, multi-node inference, or expert parallelism.
Strong communication and collaboration skills across research, kernel, infrastructure, product, and customer teams. Applicants must be authorized to work in Switzerland and provide proof of employment eligibility.
Nice to have
Experience with quantization methods and formats including FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, or SmoothQuant.
Experience with distillation, speculative decoding, EAGLE, Medusa, or multi-token prediction.
Experience supporting agentic workloads with tool calling, structured outputs, streaming APIs, and high concurrency.
Familiarity with CUDA, Triton, NCCL, NVLink, InfiniBand, or RoCE.
Open-source contributions to projects such as vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, or KServe.
Culture & Benefits
Competitive compensation.
Career growth and learning opportunities.
Flexibility, ownership, and a fast-moving environment.
Collaborative and innovative international culture.
Opportunity to work on impactful AI projects with experienced engineering teams.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →