Description
Mindrift is recruiting experienced software developers for a project-based opportunity to create and evaluate benchmark tasks for AI coding agents. The role involves building realistic simulated developer environments, designing solvable coding tasks, writing functional and integration tests, reviewing agent-generated solutions, and refining evaluations based on QA feedback. It is project-based rather than permanent employment, with tasks estimated at about 30 hours and compensation paid per accepted task, up to the equivalent of $200 per hour.
