Turing is seeking experienced, hands-on software engineers to evaluate and improve AI coding models. The work focuses on collaborating with coding agents across real-world software repositories and assessing the quality of their work, rather than primarily building production applications. The engineer reviews generated code and agent behavior, determines whether solutions are technically correct, identifies failure modes, and creates evaluation signals and feedback to improve model performance. The role includes developing evaluation and preference data, creating and refining rubrics, and building and maintaining infrastructure for data generation, collection, and evaluation workflows. The engineer will work closely with researchers and engineers to translate qualitative technical judgment into scalable processes, and communicate findings through clear write-ups, updates, and recommendations. This is an independent contractor engagement expected to last approximately three months. It is open to candidates in North America, LATAM, or India. The preferred availability is 40 hours per week, with at least six hours of Pacific Time overlap. The start is as soon as possible. Compensation is market rate, and candidates are asked to provide a specific hourly rate expectation. The evaluation process consists of an approximately 25-minute AI interview, a 30-minute practical code/AI evaluation exercise, and an approximately 20-minute hiring manager interview. The practical exercise assesses the ability to review and evaluate AI-generated code, not competitive programming or algorithm puzzles.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
Career Level
Senior
Education Level
No Education Listed