RL Environments Engineer

What you'll do

• Build environment-generation pipelines with templating, automated grading, verification, and QA.
• Create realistic coding environments and scale production to thousands of validated agentic tasks.
• Own task iteration and quality, including testing with frontier models, failure analysis, and preventing reward hacking or grader loopholes.

What they ask for

• Demonstrated experience shipping agentic coding tasks or environments at volume, including personal output and production cost.
• Experience scaling task production through automation.
• Strong software engineering fundamentals and fluency in several programming languages.
• Production software experience with large codebases, build systems, testing, deployment, on-call, and root cause analysis.

Nice to have

• Experience with RL training systems, post-training, verifiers, or tool-use harnesses; developer tooling, CI/CD, sandboxes, or code execution infrastructure.
• Experience with automated test generation, fuzzing, or benchmark suites; contributions to public agentic benchmarks or relied-upon open-source projects.

Pay

$250,000 – $300,000 USD / year (base salary)

Tools

RL, CI/CD, Terminal-Bench

AI focus

Build and validate RL environments and agentic coding tasks using frontier coding agents and models; analyze failures and improve task quality.