AI Trainer / Quality Reviewer (DataAnnotation Tech)
Evaluated AI-generated responses in production settings by applying detailed rubrics and adapting to changing project guidelines. Conducted multi-turn LLM testing to assess behavior against specific goals, then reported bugs and edge cases with concise written rationale. Performed claim verification and accuracy checks by researching and fact-checking information, followed by editing for clarity and instruction-following. • Tested multi-turn conversations and specific task goals • Compared, ranked, and critiqued responses using rubrics • Researched/fact-checked claims for accuracy and relevance • Identified and reported bugs, unexpected behavior, and edge cases