AI Interaction Evaluator (Codex / Claude Code, up to $200/hr)

📋 Description Evaluate AI-generated coding interactions end-to-end Judge usefulness, preamble and reasoning quality Assess output for strong engineering judgment Evaluate interaction feel for an experienced developer This role emphasizes engineering taste over syntax correctness Hours: ~10–20 hrs/week; duration ongoing; take-home exercise and interview 🎯 Requirements Staff / Principal-level engineer (or equivalent) Strong background in TypeScript/JavaScript Python Hands-on with OpenAI Codex, Claude Code, Cursor Deep familiarity with AI-assisted dev workflows Able to evaluate code without fully executing or reviewing every line Direct, opinionated feedback; high bar for good engineering