Research Engineer, Code RL (Reinforcement Learning)

Overview

Research Engineer on Anthropic’s Code RL team, advancing models’ ability to write, edit, test, debug, and ship real software end to end on real codebases.

What you'll do

• Advance model performance for code-writing, editing, testing, debugging, and shipping on real codebases using real tools.
• Design RL environments and coding tasks.
• Build reward signals and verifiers for what “good code” means.
• Run training experiments on frontier models.
• Diagnose why models do or do not improve on software-engineering tasks.
• Improve speed and reliability of training and evaluation pipelines.

What you'll need

• Strong software-engineering skills and deep Python expertise, including async/concurrent programming.
• Comfort owning systems end to end and debugging across the stack.
• Ability to balance research exploration with engineering implementation and shape experimental design and interpretation.
• Care about code quality, testing, and performance.
• Experience or familiarity with reinforcement learning concepts such as RLHF or post-training (stated as a strong-candidate item).
• Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience (field relevant to the role).

Details

• Annual compensation range: $500,000 to $850,000 USD.
• Hybrid policy: expect staff in an office at least 25% of the time.