Full Stack Software Engineer, Reinforcement Learning

Anthropic's Reinforcement Learning organization leads the research and development that trains Claude to be capable, reliable, and safe. We've contributed to every Claude model, with significant impacts on the autonomy and coding capabilities of our most advanced models. Our work spans developing systems that enable models to use computers effectively, advancing code generation through reinforcement learning, pioneering fundamental RL research for large language models, and building scalable training methodologies.

The RL org is organized around four key goals: solving the science of long-horizon tasks and continual learning, scaling RL data and environments to be comprehensive and diverse, automating software engineering end-to-end, and training the frontier production model. We collaborate closely with Anthropic's alignment and safety teams to ensure our systems are both capable and safe.

Our engineering teams build the environments, evaluation systems, data pipelines, and tooling that make all of this possible: from realistic agentic training environments and scalable code data generation, to human data collection platforms and production training operations.

About the Role

As a Full-Stack Software Engineer within Reinforcement Learning, you'll build the platforms, tools, and interfaces that power RL environment creation, data collection, and training observability. Our ability to train frontier models depends on generating diverse, high-quality training data and the products you build are what make that possible for researchers, vendors, and data labelers alike.

This is a software engineering role embedded within research teams. You'll own product surfaces end-to-end from backend services and APIs to web UIs that internal researchers, external vendors, and data labelers rely on daily. You don't need a background in ML research what matters is strong full-stack engineering skills and the ability to build polished, reliable products in a fast-moving environment.

You May Be a Good Fit If You:

• Have strong software engineering fundamentals with full-stack experience
• Are proficient in Python and modern web frameworks (React, TypeScript, or similar)
• Have experience building and shipping user-facing products, internal tools, or developer platforms
• Can own a product surface end-to-end backend, frontend, API design, database schema
• Have experience with relational databases, API design patterns, and authentication/authorization systems
• Care about UX and can build interfaces that are intuitive for both technical and non-technical users
• Communicate clearly with researchers, operations teams, and engineers and can translate ambiguous requirements into well-scoped work
• Are motivated by building excellent platforms
• Operate with high agency: you identify what needs to be done and drive it forward independently
• Thrive in a fast-paced environment where priorities shift and new problems emerge regularly
• Care about Anthropic's mission to build safe, beneficial AI and want your work to contribute to that goal

Strong Candidates May Also Have:

• Experience building data collection, labeling, or annotation platforms
• Background building multi-tenant platforms with role-based access and vendor management workflows
• Experience with cloud infrastructure (GCP or AWS), Docker, and CI/CD pipelines
• Familiarity with LLM training, fine-tuning, or evaluation workflows
• Experience with async Python frameworks (Trio, asyncio) or high-throughput API design
• Background building dashboards, monitoring, or observability tooling
• Experience working with external vendors or partners on technical integrations

Representative Projects:

• Building a unified platform for human data collection that integrates labeling workflows, vendor management, and quality assurance for complex agentic tasks
• Developing vendor onboarding automation that handles Docker registry access, API token management, and environment validation
• Creating evaluation and observability dashboards that catch reward hacks, measure environment difficulty, and provide real-time feedback during production training
• Building environment quality review workflows that allow researchers to browse, grade, and provide feedback on training environments
• Developing automated environment quality pipelines that validate correctness and difficulty calibration before deployment to production training
• Building internal tools for browsing and analyzing training run results, environment statistics, and data collection progress

#J-18808-Ljbffr