Medical AI optimization and RLHF medical expert (training/response evaluation)
Provided medical AI training support focused on RLHF-style feedback and response quality improvement for healthcare LLMs. Performed medical fact-checking and “hallucination” detection for clinical-style medical text to reduce safety risks. Supported evaluation of helpfulness and tone of AI responses for professional and patient-facing contexts. • Identified inaccuracies in AI-generated medical advice using clinical knowledge • Verified outputs against international medical terminology and guidelines • Assessed response helpfulness and tone for user intent alignment • Delivered clinically safety-focused feedback for global healthcare AI models