The role
Work directly with the founders to turn demanding workflows into reliable training environments and datasets. You’ll own systems from the first task prototype through production, using coding agents as part of your everyday engineering practice.
What you’ll work on
- Create multi-step coding tasks, isolated runtimes, test suites, and reference solutions.
- Build pipelines for task generation, environment orchestration, evaluation, and quality control.
- Translate researchers’ needs into repeatable workflows across repositories, terminals, APIs, and tools.
- Take ownership of a core product or infrastructure area and ship it end to end.
What you bring
- 3+ years of professional software engineering experience.
- Strong Python and practical experience with TypeScript, Docker, Git, APIs, and databases.
- A track record of using coding agents to build and debug substantial software.
- Care for reproducibility, well-designed tests, and the details that make data useful.
Show us your work
Show us a system you shipped, an open-source contribution, or an agent workflow that made a difficult engineering task easier.