Medical Query System AI/LLM Trainer & Evaluator
OtherTextText Generation
Designed and implemented a Retrieval-Augmented Generation (RAG) pipeline for answering clinical and genetic disorder queries using OpenAI’s LLMs. Developed, trained, and evaluated models on custom datasets sourced from biomedical texts to ensure high-quality, context-relevant responses. Built evaluation datasets and workflows for rigorous assessment of model retrieval and generation accuracy.
• Conducted prompt engineering and created varied prompt/response pairs for supervised fine-tuning.
• Curated, split, and labeled biomedical textual datasets for question answering.
• Evaluated outputs for factuality, completeness, and clinical relevance.
• Integrated model with domain-specific metrics for robust assessment.
Designed and implemented a Retrieval-Augmented Generation (RAG) pipeline for answering clinical and genetic disorder queries using OpenAI’s LLMs. Developed, trained, and evaluated models on custom datasets sourced from biomedical texts to ensure high-quality, context-relevant responses. Built evaluation datasets and workflows for rigorous assessment of model retrieval and generation accuracy. • Conducted prompt engineering and created varied prompt/response pairs for supervised fine-tuning. • Curated, split, and labeled biomedical textual datasets for question answering. • Evaluated outputs for factuality, completeness, and clinical relevance. • Integrated model with domain-specific metrics for robust assessment.
Not specified