Text Annotation & Classification, Content Quality & AI Response Evaluation
Labeled text data for categories such as topic classification, sentiment, toxicity/safety, and content appropriateness, ensuring consistency across large datasets. Reviewed AI-generated responses for accuracy, helpfulness, tone, and policy compliance. Provided structured feedback and ratings to help improve model performance.