Mindrift · United States
Join Mindrift to build a cutting-edge dataset for evaluating AI coding agents. You’ll design realistic developer environments, craft challenging tasks, and write robust tests to measure how well AI models handle real-world coding scenarios. This is not data labeling or prompt engineering — it’s about understanding AI limitations and creating fair, high-impact evaluation benchmarks.
About the Role
Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. This role is project-based, not permanent employment. You’ll work remotely to create a dataset that evaluates AI coding agents — how well they handle real-world developer tasks. You’ll build simulated environments, design tasks from intermediate states, and write tests that accept all valid solutions while rejecting incorrect ones. The goal is to create a fair, robust evaluation system that reveals where AI models succeed or fail.
What You Will Do
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
What We Are Looking For
8+ years in software development
Core stack: Python (FastAPI), JavaScript/TypeScript (React), Docker, Postgres, Kafka, Redis
Experience writing tests (functional, integration)
English proficiency – B2+
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.
Why this is hard: 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.
How it works: Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid. Tasks are estimated to take 30 hours to complete, depending on complexity. This is an estimate and not a schedule requirement; you choose when and how to work. Tasks must be submitted by the deadline and meet the listed acceptance criteria to be accepted.
Compensation: Up to $200/hr equivalent, depending on level and pace. Tasks are estimated at ~30 hours each; you set your own schedule.
Apply for this position now — your expertise can shape the future of AI evaluation.