Healthcare/Medicine AI Fellow
TextRLHF
As a Healthcare/Medicine AI Fellow, I annotated and validated over 12,500 clinical and biomedical prompts for LLM training. I designed chain-of-thought medical explanations and improved model accuracy on standardized medical reasoning tasks. I performed response ranking and flagged hallucinations in AI outputs to enhance clinical safety. • Annotated LLM prompts and biomedical texts for AI model training
• Performed RLHF response ranking and guideline alignment for clinical outputs
• Contributed to prompt engineering for multi-modal medical reasoning (text and image)
• Collaborated on refining annotation protocols, raising inter-rater agreement to best-in-class standards.
As a Healthcare/Medicine AI Fellow, I annotated and validated over 12,500 clinical and biomedical prompts for LLM training. I designed chain-of-thought medical explanations and improved model accuracy on standardized medical reasoning tasks. I performed response ranking and flagged hallucinations in AI outputs to enhance clinical safety. • Annotated LLM prompts and biomedical texts for AI model training • Performed RLHF response ranking and guideline alignment for clinical outputs • Contributed to prompt engineering for multi-modal medical reasoning (text and image) • Collaborated on refining annotation protocols, raising inter-rater agreement to best-in-class standards.
2025 - Present