This project aims to develop and validate an AI-driven data extraction pipeline using Large Language Models (LLMs) to transform unstructured clinical notes into structured data within hospital electronic medical records (EMRs). The student will use a high-performance computing (HPC4Health) environment to locally deploy pre-trained LLMs, build data ingestion and output processes, and use statistical methods to evaluate accuracy. The focus will be on developing quantitative features/variables related to social determinants of health and technology use among hospitalized children. This work will advance the use of AI methods in healthcare data science and inform quality improvement and research in pediatric care. The student will be embedded in the lab of Dr. Mahant, SickKids Research Institute , and co-supervised by Professor Nathan Taback, Department of Statistical Sciences, Faculty of Arts & Science University of Toronto. The SUDS Scholarwill gain hands-on experience in: High-performance computing and data science workflows; Deploying and fine-tuning large language models; Clinical informatics and healthcare data systems; Statistical methods to evaluate and compare data accuracy and ethical data stewardship; and, Applying AI and data science techniques to real-world clinical research questions.