AI Training Data Specialist / Annotator — Outlier AI
Performed AI response evaluation and quality rating against strict rubric criteria for reasoning, logic, and tone. Identified and reported errors to improve model accuracy and safety. Contributed to RLHF-style response ranking and fact-checking quality assurance for reliable deployments.• Evaluate and rate AI-generated outputs using defined quality criteria.• Detect reasoning, logic, and tone issues in responses.• Conduct hallucination detection by flagging factual inaccuracies.• Support prompt refinement and response ranking consistency with guidelines.