Research Engineer, Machine Learning (Reinforcement Learning)

Anthropic's Reinforcement Learning team seeks a Research Engineer to advance AI capabilities and safety. The role involves developing RL infrastructure, training methodologies, and agentic models for tasks like computer use and code generation. You'll collaborate on cutting-edge research and engineering to improve LLM reasoning and autonomy.

• Architect and optimize core reinforcement learning infrastructure, from clean training abstractions to distributed experiment management across GPU clusters. Design, implement, and test novel training environments, evaluations, and methodologies for reinforcement learning agents. Drive performance improvements through profiling, optimization, and benchmarking. Implement efficient caching solutions and debug distributed systems.

• Proficiency in Python and async/concurrent programming (e.g., Trio), experience with ML frameworks (PyTorch, TensorFlow, JAX), industry experience in ML research, ability to balance research and engineering, strong systems design and communication skills.

• Familiarity with LLM architectures and training methodologies, experience with RL techniques and environments, virtualization, Kubernetes, distributed systems or HPC, Rust and/or C++.