Research Engineer – Machine Learning (Reinforcement Learning)

About the role

Anthropic is seeking a Research Engineer to advance reinforcement learning methods for large language models. You will work at the intersection of cutting‑edge research and scalable engineering, helping to build safer, more capable AI systems.

Key responsibilities

• Architect and optimise core reinforcement‑learning infrastructure, from clean training abstractions to distributed experiment management on GPU clusters.
• Design, implement, and evaluate novel training environments and methodologies that push the state of the art for agentic models.
• Drive performance improvements across the stack through profiling, optimisation, benchmarking and efficient caching solutions.
• Collaborate with alignment, red‑team, and production training groups to translate research innovations into deployed models.
• Develop automated testing frameworks, clean APIs and robust evaluation pipelines.

Required profile

• Strong research mindset with a proven ability to translate ideas into production‑ready systems.
• Experience building large‑scale machine‑learning pipelines and working with distributed GPU resources.
• Commitment to AI safety and collaborative, cross‑functional teamwork.

Required skills

• Reinforcement Learning
• Distributed Systems
• GPU Computing
• Experiment Management
• Profiling and Performance Optimisation
• API Design