Freelance AI Trainer & Prompt Evaluator
As a freelance AI Trainer & Prompt Evaluator, I provided structured human feedback for reinforcement learning from human feedback (RLHF) pipelines. My tasks included reviewing and rating AI-generated responses, crafting prompts, annotating datasets, and evaluating AI-generated outputs for quality and factual accuracy. The work contributed to model alignment and robustness by improving response helpfulness, safety, and instruction-following. • Reviewed and rated AI-generated text for quality, safety, and instruction adherence. • Created and submitted diverse prompts to test large language models. • Annotated text datasets for tasks such as NER, sentiment analysis, and classification. • Evaluated code, creative writing, and factual Q&A for consistency and accuracy.