Freelance AI Trainer & Data Annotation Specialist
I trained large language and multimodal models on diverse data sources, spanning coding, general knowledge, creative writing, law, and science topics. My work involved labeling and rating AI responses, performing reinforcement learning from human feedback, and red-teaming for edge-case scenario detection. I contributed heavily to data annotation tasks including entity recognition, intent labeling, sentiment analysis, and prompt or response generation. • Wrote and reviewed a large volume of prompts and responses for instruction tuning and RLHF. • Annotated text, images, and audio across a wide range of labeling tasks. • Evaluated model accuracy by identifying errors, bias, hallucinations, and surface-level inaccuracies. • Participated in guideline development and calibration sessions to standardize annotation quality.