Nebius Token Factory is building an AI training and model post-training capability for frontier model improvement. This role owns the infrastructure that makes large-scale training and RL experiments possible, reliable, reproducible, and efficient. The work sits at the intersection of distributed systems, GPU performance, model training frameworks, RL pipelines, and production engineering
A Senior Machine Learning Engineer owns substantial ML work end to end. They can translate an ambiguous capability goal into concrete experiments, implement and debug training and RL recipes, build the supporting data and systems, and deliver measurable improvements in model quality, experiment throughput, and reliability. They are deeply hands-on and can independently debug both model-behavior failures and distributed training failures
Design and run model-training and post-training experiments, including SFT, continued pretraining, preference optimization (DPO/IPO/KTO), and RL methods such as RLHF/RLAIF, PPO, and GRPO
Build reward functions, judge models, verifiers, task environments, and evaluation sets for reasoning, coding, tool use, and agentic workflows
Create synthetic data and data pipelines, including teacher-student generation, self-play, rejection sampling, filtering, and quality scoring
Analyze model-behavior failures and turn them into targeted data, reward, or algorithm improvements
Build and maintain distributed training and RL infrastructure using frameworks such as Megatron-LM, DeepSpeed, PyTorch FSDP/DTensor, Ray, verl, slime, AReaL, or OpenRLHF
Implement and debug parallelism strategies (tensor, pipeline, sequence/context, expert, and data parallelism) and build reliable rollout, reward-serving, checkpointing, and experiment-orchestration components
Profile and improve GPU utilization, memory usage, communication efficiency, training throughput, and inference/serving performance
Design rigorous evaluations and ablations for capability, instruction following, reasoning, tool use, safety, and regression risk
Write clear experiment plans, design docs, benchmark reports, and runbooks, and partner across research and platform teams