Senior Python Engineer — AI Evaluation & Testing Architect
Summary
A project-based role at Mindrift building datasets to evaluate AI coding agents: you create realistic developer environments, design tasks from intermediate states, craft prompts, and define test criteria that accept valid solutions and reject incorrect ones. Core tech is Python, with 8+ years of software development experience required.
Mindrift is building a dataset to evaluate AI coding agents by creating realistic developer environments—tickets, docs, and a believable history of code changes. You’ll design tasks from intermediate states, craft prompts, and define what counts as a solved task, ensuring tests accept all valid approaches and reject incorrect ones.
This project-based role requires 8+ years in software development and English proficiency (B2+).