Research Engineer, RL Engineering

About the role

Anthropic is looking for a Research Engineer to join its Reinforcement Learning Engineering team. You will build and improve the systems that power the training of large language models such as Claude, enabling rapid advances in AI capability and safety.

Key responsibilities

• Design, implement, and maintain high‑performance algorithms and infrastructure for RLHF and related finetuning methods.
• Profile and optimise the reinforcement‑learning pipeline to increase speed and reliability.
• Develop tooling that automatically launches training jobs in test environments for early problem detection.
• Adapt finetuning systems to new model architectures and resolve performance regressions.
• Instrument code to eliminate Python GIL contention and other bottlenecks.

Required profile

• 4+ years of software engineering experience, preferably in ML systems.
• Passion for building tools that make researchers more productive.
• Result‑oriented mindset with flexibility to tackle tasks beyond the core description.
• Enjoys pair programming and collaborative problem solving.
• Interest in the societal impact of AI technologies.

Required skills

• Python programming.
• High‑performance, large‑scale distributed systems.
• Large‑scale LLM training experience.
• Implementation of LLM finetuning algorithms such as RLHF.