Daniel Palacios is a PhD student in Quantitative and Computational Biosciences whose research focuses on developing agentic AI systems for clinical outcomes prediction. His work combines foundation models, automated machine learning, and rigorous statistical validation to extract actionable insights from electronic health records at scale.
His research spans postpartum mental health prediction, pediatric epilepsy and diabetes analysis, and rare disease phenotyping across 50,000+ patients. He developed ClinPreAI, an agentic AI system for postpartum depression risk prediction using multimodal EHR data, which has resulted in publications and awards including a BCM Seed Grant and dkNET NIDDK Grant. Daniel has extensive experience handling large-scale medical datasets, and performing data curation across AWS and Palantir Foundry platforms. His technical work emphasizes the practical deployment of AI systems with clinical validation.
Prior to his doctoral work, Daniel contributed to NASA space biology research, developing the AMMPER agent-based model for simulating microbial radiation exposure and building NLP taxonomy systems for space hardware failure analysis at NASA Johnson Space Center. His work has been disseminated through publications and 20+ conference presentations, with open-source software contributions spanning space biology and clinical informatics.
As an NSF Graduate Research Fellow and NLM Fellow, Daniel continues to advance the application of AI to healthcare’s most challenging data integration and prediction problems, bridging computational innovation with clinical deployment.