OutlierSenior software engineer — RLHF for code models
Help frontier labs train better code models by writing reference solutions and grading model output.
$55–80/hrRemote — WorldwidecontractApr 23, 2026
You will write reference implementations in Python, TypeScript, and one or more systems languages, then evaluate model-generated solutions against them. Most projects are async with weekly deliverable targets.
This role is best suited to engineers with 5+ years of professional experience and a strong intuition for code review.
Requirements
- 5+ years professional software engineering experience
- Strong fluency in Python and one of: TypeScript, Go, Rust, C++
- Experience giving detailed code review feedback