Staff Software Engineer, Code RL
San Francisco, CA | New York City, NY | Seattle, WA
Posted 6d ago
Job Location
San Francisco, CA | New York City, NY | Seattle, WA
Tech Stack
Remote Work Policy
On-site
Categories
Applied AI Engineer
About the job
Anthropic is building reliable, interpretable, and steerable AI systems to be safe and beneficial for users and society. This role focuses on the engineering aspects of reinforcement learning for Claude's coding capabilities, involving the creation and scaling of agentic coding environments. You will have significant influence on technical direction and standards, embedding with research teams to understand their needs, build supporting frameworks and infrastructure, and then transfer ownership of these well-maintained systems. The role also includes ensuring the ongoing health and maintainability of production RL runs, including monitoring and triage tooling. The team's work spans client-side sandboxed execution for agentic RL, large-scale data processing, dataset lifecycle management, and the frameworks researchers use to build environments. You will focus on areas where your deep expertise is most valuable, particularly if you have strong Python skills, a keen eye for API and framework design, and experience with complex system failures.
Responsibilities
- Design widely-used APIs, frameworks, and abstractions with attention to interface legibility and principled defaults.
- Embed with research teams to build systems and APIs supporting their work, and transfer ownership.
- Improve reliability and structure in research codebases without hindering research progress.
- Prevent silent failure modes through system design, type safety, and targeted testing.
- Contribute to the reliability of production RL systems, including monitoring and triage tooling.
- Help define engineering standards, review practices, and design patterns for a new team.
- Mentor researchers and engineers in adopting new standards and practices.
Requirements
- Deep expertise in Python, including static typing, safe async/concurrency, and performant code.
- Proven track record of designing adopted APIs or frameworks.
- Experience working productively in large, evolving, or research-style codebases.
- Demonstrated ability to anticipate and prevent silent failure modes structurally.
- Strong written and verbal communication skills, able to explain system designs to diverse audiences.
- Comfort with ambiguity and ability to scope and drive loosely defined problems to maintainable outcomes.