Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. This role involves building a dataset to evaluate AI coding agents by creating realistic developer environments, designing tasks, and writing tests to verify agent solutions. Participation is project-based, not permanent employment.
Responsibilities
- Build realistic developer environments - a virtual company with codebase, infrastructure, and context (tickets, docs, conversations) that forms a believable development history
- 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
- Write tests that verify agent solutions - accept all valid approaches and reject incorrect ones, neither too strict nor too lenient
- Iterate on tasks and tests based on QA feedback - review agent solutions, analyze failures, and refine until the evaluation is fair and robust
- Submit tasks by the deadline and meet the listed acceptance criteria to be accepted
Requirements
- 5+ years in software development
- Core stack: Python (FastAPI), JavaScript/TypeScript (React), Docker, Postgres, Kafka, Redis
- Experience writing tests (functional, integration)
- English proficiency - B2+
- Submit CV in English and indicate level of English proficiency
Benefits
- Compensation up to $30/hr equivalent, depending on level and pace
- Flexible schedule - you choose when and how to work
Additional details
- Participation is project-based, not permanent employment
- This is NOT data labeling
- This is NOT prompt engineering
- This is NOT writing code from scratch - the agent writes most of the code; you guide and evaluate
- 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.
- Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid
- Tasks for this project are estimated to take 20 hours to complete, depending on complexity. This is an estimate and not a schedule requirement
- Originally posted on Himalayas
Zobhira
