Research Engineer - Distributed Training

What this role involves

Prime Intellect is hiring a Research Engineer focused on distributed training. The role involves building and optimizing distributed training infrastructure for large-scale pre-training and RL workloads, contributing to open-source frameworks like prime-rl. The ideal candidate has strong systems engineering experience in AI/ML infrastructure, deep familiarity with PyTorch and distributed training frameworks, and experience optimizing training performance.

Skills this posting asks for

• pytorch
• pytorch distributed
• deepspeed
• fsdp
• megatron
• vllm
• ray
• cuda
• triton
• distributed training
• data parallelism
• tensor parallelism
• pipeline parallelism
• gpu architecture
• performance profiling
• kernel optimization
• communication optimization
• memory optimization
• rl training
• async rollout
• compiler optimization
• runtime optimization
• multi-node gpu clusters
• high-performance networking

From the employer’s posting

OWN YOUR INTELLIGENCE Prime Intellect is building the open superintelligence stack: the infrastructure frontier AI labs build internally, made available to every ambitious AI team. Our platform, Lab, unifies compute, environments, evaluations, secure sandboxes, high-performance training, and…