Mindrift is building datasets to evaluate AI coding agents, working with project-based engagements.
You will craft challenging tasks within simulated environments, define what counts as solved, and write tests that accept multiple correct approaches.
This role emphasizes evaluating and refining tasks based on QA feedback.
We seek experienced Senior Python Engineers with strong FastAPI, pytest, Docker, PostgreSQL, and CI/CD skills, plus English proficiency and 30+ hours weekly availability.