The growing complexity and volume of biomedical research data, especially in infectious and immune mediated diseases, calls for novel approaches to data enrichment and information extraction. A key challenge is efficiently creating FAIR compliant metadata from unstructured clinical documents, a process currently requiring significant time and expertise. This project aims to advance this field by developing automated AI models for metadata enrichment and standardization, leveraging John Snow Labs’ Medical Large Language Models (LLM) and Natural Language Processing (NLP) technology. This includes models that will deliver state-of-the-art accuracy, on benchmarks validated by medical doctors, on a productiongrade codebase that scales natively on commodity hardware. New models will be trained for Named Entity Recognition (NER) and Entity Resolution, that extract and standardize unstructured data into multiple biomedical ontologies. This research will bring transformative improvements in biomedical data management for infectious and immune-mediated diseases, significantly streamlining metadata enrichment processes, enhancing data utility for researchers and streamlining scientific data sharing.