Please submit your CV in English and indicate your level of English proficiency.<\/h4>\n

Mindrift<\/strong> connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.<\/p>\n

What this opportunity involves<\/h4>\n

We're building a dataset to evaluate AI coding agents - how well a model handles real-world developer tasks.<\/p>\n

You'll create challenging tasks and evaluation criteria within realistic simulated environments:<\/p>\n

\n
• Build realistic developer environments - a virtual company with codebase, infrastructure, and context (tickets, docs, conversations) that forms a believable development history<\/li>\n
• Design tasks from intermediate states of these environments - craft the prompt, define what \"solved\" means, and ensure the task is solvable by an AI agent<\/li>\n
• Write tests that verify agent solutions - accept all valid approaches and reject incorrect ones, neither too strict nor too lenient<\/li>\n
• Iterate on tasks and tests based on QA feedback - review agent solutions, analyze failures, and refine until the evaluation is fair and robust<\/li>\n<\/ul>\n

  What this is not<\/h4>\n

  \n
  • Not data labeling<\/li>\n
  • Not prompt engineering<\/li>\n
  • Not writing code from scratch - the agent writes most of the code; you guide and evaluate<\/li>\n<\/ul>\n

    What we look for<\/h4>\n

    \n
    • 5+ years in software development<\/li>\n
    • Core stack: Python (FastAPI), JavaScript\/TypeScript (React), Docker, Postgres, Kafka, Redis<\/li>\n
    • Experience writing tests (functional, integration)<\/li>\n
    • English proficiency - B2+<\/li>\n<\/ul>\n

      Why this is hard<\/h4>\n

      Frontier models are already good at coding. Creating a task that genuinely challenges the best models is non-trivial. You need to deeply understand where models fail and what scenarios reveal the difference between a good and a bad solution. Tasks have many valid solutions - writing tests that accept all correct solutions and reject incorrect ones is harder than it sounds.<\/p>\n

      How it works<\/h4>\n

      Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.<\/p>\n

      Compensation<\/h4>\n

      Up to $40\/hr equivalent, depending on level and pace. Tasks are estimated at approximately 20 hours each; you set your own schedule.<\/p><\/p>

      <\/p>