Generative AI Subject Matter Expert (SME): Biomedical LLM SFT, RLHF, and Evaluative Data Annotation
As a Generative AI Subject Matter Expert, I engineered and refined specialized prompts for the training, evaluation, and benchmarking of large language models (LLMs) focused on biomedical and scientific topics. I performed quality assurance on data annotations, reviewed and edited AI-generated responses, and authored expert-level 'gold standard' data for model training. My work involved testing, evaluating, and iteratively refining model outputs in collaboration with a multidisciplinary team. • Developed and edited prompt-response pairs for SFT and RLHF in scientific domains. • Conducted quality assurance and provided detailed feedback on LLM output accuracy. • Benchmarked model performance using expert-authored responses across diverse topics. • Collaborated extensively with AI professionals and other subject-matter experts.