AI Response Evaluation and Data Annotation
Created and annotated high quality text datasets for training NLP models. Rated and ranked AI generated responses on accuracy, truthfulness, relevance, safety and followed instructions. Compared different model outputs, found areas for improvement, and gave detailed human feedback to help develop and align AI systems.