Machine learning
Foundation models, agentic reasoning and predictive systems built to be validated, not just demoed.
We build AI for health and life sciences, and we organize our work as a set of independent projects, each built around a specific scientific or clinical problem.
Drag a molecule →
Every project starts from a specific scientific or clinical question, not a technology in search of a use.
We build models and systems to hold up under scrutiny and real testing, not just to look good in a demo.
We turn the research into diagnostic tools, devices and software that work in real clinical settings.
Foundation models, agentic reasoning and predictive systems built to be validated, not just demoed.
Oncology, cardiovascular disease, sleep, neurodegeneration and more, grounded in real clinical questions.
Turning models into diagnostic and decision-support tools that hold up in real clinical practice.
Applied AI across the full arc of health — from prediction and prevention to diagnostics, imaging, therapeutics and care.
Projects across the center, from digital twins to clinical AI.