MAY 29, 2026
Subigya Nepal, PhD
Assistant Professor, Computer Science
University of Virginia
About the Presentation: Our wellbeing unfolds in the rhythms of daily life, in how we sleep, move, talk and spend our time, yet our ability to understand and support it has long been limited by what clinicians can observe and what people can recall and report. Passive sensing from smartphones and wearables offers a continuous, unobtrusive window into these behaviors, and large language models now make it possible to turn those signals into something a person or a clinician, can actually use.
In this talk, Dr. Nepal presents a line of research that pairs behavioral sensing with AI to understand and support wellbeing across real-world and clinical populations. He will discuss how grounding AI in a person’s behavioral context can deepen self-reflection and improve mental-health outcomes, what rigorous randomized evaluation reveals about when AI-mediated support genuinely helps and when it does not and how emerging agentic systems can investigate behavioral data to anticipate the moments people most need support. Together, this work points toward just-in-time, personalized intervention and raises practical questions about how such systems should be built, validated and responsibly deployed in behavioral health.
About the Presenter: Dr. Subigya Nepal is an Assistant Professor of Computer Science at the University of Virginia, where he directs the SONDER Lab (Sensing, Observing, aNd unDerstanding human ExpeRience). His research sits at the intersection of artificial intelligence, digital health and human-computer interaction. He studies how the digital traces of everyday life, from smartphones, wearables and laptops to social media and online communities, can illuminate human behavior and mental health as it actually unfolds. A second thread asks how AI can turn those signals into timely, personalized support. He earned his PhD in Computer Science at Dartmouth College with Andrew Campbell, where he led some of the longest-running mobile-sensing studies of students, workers and clinical populations. Prior to joining UVA, he was a postdoctoral fellow at Stanford’s Institute for Human-Centered AI, mentored by Gabriella Harari. His work has received Distinguished Paper Awards at UbiComp 2023 and 2025.