Freelance AI Trainer | Various AI Platforms (Remote) | 2023–Present
Annotated and evaluated AI-generated responses across scientific, medical, and general knowledge domains to improve model accuracy and safety. Crafted and refined LLM prompts while assessing relevance, coherence, factual accuracy, and tone of outputs. Produced detailed preference rankings and written rationales for RLHF pipelines, and documented edge cases, biases, and hallucinations. • Perform response annotation and evaluation across multiple knowledge domains • Create and refine prompts for LLM quality assessment • Provide preference rankings with written rationale for RLHF • Identify and report edge cases, biases, and hallucinations for research teams