Research Engineer, RL Frontiers
Core
Train proprietary frontier models to automate kernel development, ASIC engineering, production pipelines, and 3-D hardware operating systems using custom simulators and environments.
Role type
Senior Research Engineer (RL & Post-training)
Builds
Frontier AI models for hardware engineering tasks
Domain
Hardware AI / ASIC Engineering / Reinforcement Learning
Deliverable
production ML models
Required skills
Post-training (SFT, RL, PPO, GRPO), PyTorch, distributed training, RL environment building, benchmark creation, systems optimization for RL loops
Preferred skills
RL frameworks (verl, OpenRLHF, TRL, prime-rl), inference engines (vLLM, SGLang), agentic multi-turn RL, model distillation, ASIC design/verification, accelerator kernels, large-scale compute infrastructure
Responsibilities
Own the post-training stack end-to-end, build RL environments with ASIC/kernel/supercomputing engineers, create custom data mixes, create benchmarks to drive productivity and avoid reward hacking
Seniority
Senior, hands-on IC
