AI Training & Data Annotation Specialist — Independent Contractor (Remote)
Text
RLHFEvaluated LLM prompt–response pairs for instruction-following accuracy, factual correctness, coherence, and safety policy adherence across thousands of training examples. Applied multi-dimensional quality rubrics to rate and rank model outputs to support Reinforcement Learning from Human Feedback (RLHF) workflows. Detected and flagged hallucinations, logical inconsistencies, and policy violations, documenting issues to reduce downstream training errors. • Performed annotation, classification, and entity-labeling tasks across NLP domains (summarization, coding, reasoning, creative writing). • Conducted structured internet research to verify claims and enrich training data with sourced information. • Maintained 95%+ accuracy and on-time delivery across concurrent projects using remote task pipelines. • Ensured traceability and consistent reporting via online dashboards and spreadsheets.