AI Training Practice (Independent) — AI Response Evaluation & Feedback (Remote)
Evaluated AI-generated responses for accuracy, relevance, and clarity using defined quality criteria. Compared multiple model outputs and selected the best-performing response based on reasoning quality, factual consistency, and usefulness. Identified factual errors, inconsistencies, and weak reasoning, then produced improved rewrites to enhance clarity and correctness. • Accuracy and relevance checks against stated or implied facts • Consistency review for internal contradictions and weak logic • Response improvement via rewriting for clarity, tone, and usefulness • Structured judgment on quality, correctness, and overall communication