AI Data Annotator & Response Evaluator (text annotation, quality review, and RLHF evaluation/ranking)
Provided AI response evaluation grounded in structured technical troubleshooting habits to assess logical gaps, inaccuracies, and implicit meaning in model outputs. Performed quality review and consistency checks on annotated responses to ensure guideline compliance and evidence-backed rationales. Produced independent evaluation rationales without relying on AI writing tools. • Evaluated responses for nuance, completeness, and root-cause reasoning quality • Applied strict attention to detail and safety-like standards for honest scoring • Produced structured, well-evidenced written rationales for each evaluation item • Followed annotation/evaluation guidelines with gap-in-reasoning workflow optimization