The National Science Foundation (NSF) awarded a $1,200,000 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) federal grant program to the University of California, San Diego (UCSD) to develop advanced foundation models for early detection of sepsis. The project aims to create novel AI algorithms, software, and systems that can process complex clinical time series data to enable accurate and timely prediction of sepsis, a life-threatening condition. Key innovations include a transformer-based model (CTSFormer) designed for clinical time series data, automated pretraining techniques, efficient finetuning methods, and interpretability enhancements. This work addresses critical gaps in existing foundation models' ability to handle irregular and incomplete clinical data. The project will run from September 2024 through August 2028 and is expected to significantly improve sepsis detection, enabling early interventions to save lives.