Staff Machine Learning Engineer, LLM Inference Optimization
careers.nebius.com
- Job written in
- English
- Location
- Zürich
- Work type
- On-site
- Type
- Full-time
Nebius is a company pioneering a new phase in cloud infrastructure tailored for the global AI sector. They are developing a comprehensive AI cloud platform that supports developers and enterprises throughout the entire AI/ML lifecycle, from data and model training to production deployment. This platform aims to eliminate the high costs and complexities associated with building large in-house AI/ML infrastructure. Nebius is publicly traded on Nasdaq (NBIS) and is headquartered in Amsterdam, with research and development centers spread across Europe, the UK, North America, and Israel. Their team consists of over 1,500 professionals, including hundreds of engineers with deep expertise in hardware, software, and AI research and development. The Staff Machine Learning Engineer, LLM Inference Optimization role at Nebius focuses on optimizing the performance of large language models (LLMs) and vision language models (VLMs). This position involves managing the entire process from model creation to deployment, ensuring that the models are fast, reliable, and cost-effective. The engineer will work on various optimization techniques, such as quantization, speculative decoding, and KV-cache optimization, to enhance the efficiency of the models. They will also develop and implement benchmarking frameworks to evaluate the performance of these models across different metrics. To excel in this role, candidates must have strong Python and PyTorch engineering skills, as well as hands-on experience with deploying or optimizing LLM, VLM, or high-throughput transformer inference systems. They should be familiar with at least one modern inference stack, such as vLLM, SGLang, TensorRT-LLM, or Triton Inference Server. A deep understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, and long-context serving, is essential. The ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs is also crucial. Strong communication skills and the ability to collaborate with various teams, including research, kernel, infrastructure, product, and customer teams, are required. Nice-to-have skills include experience with quantization-aware training, post-training quantization, and various inference acceleration methods. Familiarity with agentic workloads, CUDA, or Triton, and contributions to open-source projects related to inference optimization are also beneficial. The role offers a dynamic and innovative work environment, with opportunities for career growth and the chance to work on impactful AI projects. Nebius values a collaborative and inclusive culture, providing a supportive environment for employees to thrive. What the role asks for: - Strong Python and PyTorch engineering skills. - Hands-on experience deploying or optimizing LLM, VLM, or high-throughput transformer inference systems. - Practical knowledge of at least one modern inference stack. - Strong understanding of transformer inference bottlenecks. - Ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. - Strong communication skills and ability to collaborate with various teams. - Experience with quantization-aware training, post-training quantization, FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, SmoothQuant, or related techniques (nice-to-have). - Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods (nice-to-have). - Experience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration (nice-to-have). - CUDA or Triton familiarity (nice-to-have). - Open-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects (nice-to-have).
Engineer: Lebenslauf-Vorlage
Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Engineer-Inserate am häufigsten nennen.
Lebenslauf-Vorlage Schweiz ansehenEngineer: was der Markt gerade verlangt
Engineer-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.
Weitersuchen
- Alle AI Engineer Jobs in der Schweiz
- Alle Jobs in Zürich
- Lebenslauf als Engineer: Beispiel und Vorlage
- Lebenslauf mit KI erstellen
- ATS-Lebenslauf prüfen
- Bewerbungsschreiben für dieses Inserat