Research Engineer, Code RL (Reinforcement Learning)
Core
Design RL environments and coding tasks, build reward signals and verifiers for code quality, run training experiments on frontier models, and improve pipelines for autonomous software engineering.
Role type
Research Engineer (Reinforcement Learning for Code Generation)
Builds
RL systems for agentic coding behaviors, code correctness, long-horizon autonomous engineering, and high-performance code for accelerators.
Domain
Artificial Intelligence / Reinforcement Learning / Software Engineering
Deliverable
production ML models
Required skills
Python (async/concurrent programming), software engineering, system design, experimental design, result interpretation, code quality assurance, performance optimization
Preferred skills
Reinforcement Learning, RLHF, LLM finetuning, program analysis, testing, verification, compilers, formal methods, PyTorch, large-scale distributed training, CUDA/GPU/TPU kernel experience, virtualization, sandboxed code execution
Technologies
Python, PyTorch, CUDA, GPU, TPU
Responsibilities
Design RL environments and coding tasks; build reward signals and verifiers; run training experiments on frontier models; diagnose model performance in software engineering tasks; improve pipeline speed and reliability.
Seniority
Mid-Senior, hands-on IC
