AI Code Reviewer & LLM Evaluator
Conducted structured evaluation of LLM-generated computer code outputs for correctness, efficiency, and code quality during hands-on professional work. Provided rankings and clear explanations as a human feedback signal to improve and evaluate AI code generation models. Worked directly with code outputs to inform RLHF and ranking-driven training data pipelines. • Evaluated code for correctness, quality, and adherence to rubric • Ranked LLM code completions against one another • Provided reasoning and human feedback for model training • Supported evaluation of production-grade outputs for model improvement
Conducted structured evaluation of LLM-generated computer code outputs for correctness, efficiency, and code quality during hands-on professional work. Provided rankings and clear explanations as a human feedback signal to improve and evaluate AI code generation models. Worked directly with code outputs to inform RLHF and ranking-driven training data pipelines. • Evaluated code for correctness, quality, and adherence to rubric • Ranked LLM code completions against one another • Provided reasoning and human feedback for model training • Supported evaluation of production-grade outputs for model improvement
2023 - Present