
Hagai Rossman
Assistant Professor of Personalized Medicine
Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)
Chief Scientist, Pheno.AI
Research
I develop data-driven computational systems that transform rich biomedical data into actionable health insights. My work integrates data-driven computational modeling and AI methods with clinical knowledge to model disease trajectories and support personalized health interventions in real-world settings.
I am a co-founder of the Human Phenotype Project, a large-scale deep phenotyping cohort integrating multi-omics, imaging, wearables, and lifestyle data.
My research agenda focuses on three interconnected themes: foundation models that learn generalizable representations from multimodal health data; knowledge-infused approaches that ground predictions in biomedical and clinical understanding; and compound AI systems that combine specialized models, agents, and large language models to support complex clinical reasoning and decision making.
I earned my PhD in Computer Science at the Weizmann Institute of Science, where I played a central role in establishing and scaling the Human Phenotype Project (HPP) and developing predictive models from nationwide electronic health records and high-dimensional phenotypic data.
Featured Publications
A foundation model for continuous glucose monitoring data A transformer-based foundation model trained on 10 million glucose measurements that generalizes across populations and predicts diverse health outcomes. Nature, 2026
Diet-microbiome associations in 10,068 individuals from the Human Phenotype Project to guide personalized nutrition A large-scale HPP analysis linking dietary patterns and gut microbiome composition to support personalized nutrition. Nature Medicine, 2026
Simulating clinical interventions with a generative multimodal model of human physiology A generative model of human physiology trained on longitudinal multimodal HPP data to simulate clinical interventions and health trajectories. arXiv, 2026