Profile

Riku Arakawa is a Ph.D. candidate at the Human-Computer Interaction Institute (HCII) in the School of Computer Science at Carnegie Mellon University, where he is advised by Mayank Goel. He builds AI assistants that sense people's activities through wearable and ambient devices, model their behavior and context in real-world settings, and provide reliable, proactive support that adapts through interaction over time. His research asks how such proactive AI assistants can remain reliable despite uncertainty about the real-world context, combining ubiquitous sensing, human–AI interaction design, and computational models of adaptation.



Two health AI assistants he developed have been deployed at four clinical sites across the US and Europe, serving more than 300 patients and families: post-operative wound care for skin cancer patients and daily activities for people living with dementia, and family-centered care for children with ADHD-related hyperactivity. He also led the industry transfer of his conversational behavior-sensing system, used by therapy and coaching companies in Japan for over six years across more than 1,000 sessions.



He has published 35+ peer-reviewed full papers (11 ACM CHI, 4 UIST, 6 IMWUT), receiving a CHI Best Paper Award and four Honorable Mention Awards. He has also received the ACM UbiComp/ISWC Gaetano Borriello Outstanding Student Award (2025) and the IEEE Pervasive Computing Emerging Rockstar (2026), was named to Forbes Asia 30 Under 30 in Healthcare & Science (2024) and MIT Technology Review's Innovators Under 35 Japan (2025), and has been supported by the Funai Foundation, Masason Foundation, Snap Research Fellowship, and Quad Fellowship.



📣 On the job market for 2026–27!
I'm seeking tenure-track faculty positions. If you know of relevant opportunities, I'd be grateful if you could reach out via email.

News

previous news

Research

My research asks: How can we build proactive AI assistants that remain reliable despite uncertainty about the real-world context? I address this question through three lenses: (1) understanding and mitigating sensing uncertainty, (2) designing uncertainty-aware human–AI interactions, and (3) enabling adaptation through everyday interactions. Together, these enable AI assistants to recognize when support is needed, decide when and how to intervene, and adapt to each user through continued use. In collaboration with clinicians and health experts, I build and deploy these assistants in real-world health and well-being contexts involving patients, caregivers, and clinicians. Currently, I am working on the projects listed below.


Honors and Awards

Research Fellowships and Grants

Secured over $500k in fellowships and research grants as the lead applicant.

earlier fellowships and grants

Academic Honors and Awards

earlier awards

Misc.

more (competitions and hackathons)

Talks

Media Coverage

Contact

rarakawa [at] cs.cmu.edu