Senior ML Engineer (Token Factory)

The Token Factory is a part of Nebius Cloud, one of the world's largest GPU clouds. We are building a high-performance inference and fine-tuning platform designed to push foundation models to their hardware limits. Our mission is to maximize throughput, minimise latency, and optimise cost-per-token across tens of thousands of GPUs. We are currently working on: Inference Optimization — identify LLM bottlenecks to drive production speedups and support a wide range of architectures; Inference engines support — implement novel speculative decoding architectures and optimize components of various LLM designs; Low Precision Training & Inference — design and productionise low-precision training and inference pipelines. Requirements include a deep understanding of ML theory and transformer architecture, experience profiling GPU workloads, knowledge of GPU memory hierarchy, and strong software engineering skills (Python). Nice to have includes experience with open-source inference engines and CUDA/C++ kernel work. Benefits & Perks include competitive compensation, growth, flexibility, collaboration, and international environment. Applicants must be authorized to work in their country of application.

The text below is the employer's original job description, extracted as written. Other details on this page, like salary, responsibilities, and requirements, are interpreted from that text by our system, not values the employer explicitly confirmed, so treat them as our best interpretation rather than verified facts.