This remote role involves reviewing and improving Python backend AI agent responses, writing detailed reasoning, and evaluating model outputs against project guidelines. Applicants should possess relevant academic or professional knowledge in AI agent backend development and be fluent in English. The position is open worldwide on a part-time contract basis requiring at least 20 hours per week.