Symbiotic
Intelligence
Support the
Next Generation

Development is continuous, but clinical care is episodic. Conditions like autism and cerebral palsy are most treatable when caught early, yet most children are diagnosed years after the first signs appear, and families in underserved regions may never be screened at all.

We develop foundation models of child development, trained on how children move, look, speak, and play, that turn everyday moments into developmental signals. These models support continuous health monitoring and care at home and in the clinic, and give pediatricians and families insight they can act on early.

Children are the most remarkable learners we know. From sparse, noisy, multimodal experience, they acquire language, common sense, and social understanding, continually and without forgetting. Today's AI systems, by contrast, are trained on adult data, aligned to adult intent, and evaluated by adult benchmarks.

We design AI agents with children's cognitive abilities: agents that learn through curiosity and social interaction, understand how children think, and adapt to each child's developmental stage. Built with psychologists, pediatricians, and parents, they are meant to grow up alongside children, with safety as the foundation rather than an afterthought.

Many rare diseases first appear in childhood, and each affects too few patients for most research programs to prioritize. Families often wait years for a diagnosis, and effective treatments remain scarce.

We pursue frontier research on pediatric rare diseases with clinical partners, using AI to bring together sparse, multimodal evidence, from clinical records and imaging to movement and behavior, so that rare conditions can be recognized sooner and understood better.