Senior Software Engineer — Agentic Coding (AI Training) | Turing | Remote (Hourly Contract)
Designed and evaluated agentic coding workflows by assessing AI-generated code outputs against production-quality criteria. Performed structured evaluation of correctness, efficiency, security, and adherence to software engineering best practices using rubrics and testing harnesses. Identified failure modes and edge cases in autonomous coding behavior to inform iterative model improvement. • Developed evaluation frameworks and rubrics for agentic system performance. • Stress-tested autonomous coding across diverse languages, architectures, and problem types. • Provided expert feedback on reasoning chains, tool usage, and output quality. • Documented breakdowns to support researcher and engineering iteration cycles.