Lead RL Infrastructure Engineer — Scalable GPU Training

AMD; Santa Clara (CA); On-site; USD 130,000 - 180,000; Full time

Benefits offered by this job

Competitive benefits package

Job summary

AMD in Santa Clara is seeking a Lead AI Research Scientist - Infrastructure Engineer to enhance reinforcement learning infrastructure. You will design distributed RL training stacks and build high-throughput systems, collaborating closely with research scientists to optimize processes.

The ideal candidate has a strong background in machine learning platforms and deep expertise in GPU orchestration. A Bachelor’s degree in Computer Science is required, with a Master’s or PhD preferred. Join a culture that values innovation and collaboration.

Qualifications

• Strong systems track record in machine learning platforms.
• Deep experience with distributed training and GPU orchestration.
• Prior ownership of reinforcement learning training infrastructure.

Responsibilities

• Design and implement distributed RL training stacks.
• Build high-throughput rollout workers and reward computation pipelines.
• Collaborate with research scientists on experiment templates and hyperparameter sweeps.
• Drive reliability through on-call rotations and runbooks.

Skills

Machine learning platforms

PyTorch or JAX

C++/Python performance tuning

GPU cluster orchestration

Education

Bachelor's degree in Computer Science

Master's or PhD preferred

Tools

NCCL/MPI-style distributed training

Job description

AMD in Santa Clara is seeking a Lead AI Research Scientist - Infrastructure Engineer to enhance reinforcement learning infrastructure. You will design distributed RL training stacks and build high-throughput systems, collaborating closely with research scientists to optimize processes.

The ideal candidate has a strong background in machine learning platforms and deep expertise in GPU orchestration. A Bachelor’s degree in Computer Science is required, with a Master’s or PhD preferred. Join a culture that values innovation and collaboration.