AI Coding Evaluator / Prompt Engineer

We're hiring experienced developers to help train and evaluate frontier AI coding models. You'll write high-quality coding prompts, review model-generated code, identify bugs, and produce clear written critiques used as training data. What You'll Do - Write challenging, well-scoped coding problems with ideal solutions - Rate and rank AI-generated code for correctness, efficiency, and style - Debug model outputs and explain failures in clear written feedback - Follow detailed rubrics for consistent evaluation quality Requirements - 5+ years of professional software development experience - Fluency in Python plus at least one of: TypeScript, Rust, Go, C++, SQL - Strong code review and technical writing skills - Comfort working independently from written guidelines - Available to work on EST - Must work on remote mac machine Nice to Have - Experience with RLHF, data labeling, or AI evaluation projects - Background in algorithms, systems programming, or security - Open-source contributions or technical blog writing