Cyber-Physical Systems
for Accessibility and Ability Augmentation

My research sits at the intersection of human-AI interaction and AI for social good. I am broadly interested in how intelligent systems can be designed to preserve human agency and critical thinking—particularly in contexts where AI outputs carry uncertainty or where users are underrepresented in default interface design. This includes questions of how to surface model confidence in ways that build rather than undermine user trust, how adaptive AI systems can serve diverse user needs without requiring explicit configuration, and how the internal dynamics of large language models (such as reasoning depth and hidden-state convergence) relate to the transparency and reliability that end-users experience.

My research interests lie in augmented reality and accessibility. I design and develop AR technology to automatically recognize surrounding visual information that cannot be easily perceived by people with low vision, and generate suitable multi-modal feedback to enhance their perceptual abilities in various daily tasks.

I work on Assistance Systems for Skill Acquisition.

I work on accessibility technology for people with visual impairments in outdoor sports. My project SnowEcho is a blind skiing assistance system that conveys obstacles and course boundaries through spatial audio, allowing skiers to perceive traversable free space and choose their own line rather than following a prescribed route.

I challenge outcome-driven "access" and develop a process-oriented frame: Developmental Flourishing. It reframes assistive technologies not as instruments for reaching task outcomes, but as a means through which blind and low-vision individuals can explore, create, and construct meaning from their experiences. I share an analysis of sensing and AI systems, discussing the extent to which they align with and augment blind people’s perceptual and cognitive processes, while highlighting persistent challenges and future directions.

I am broadly interested in Artificial Intelligence and Human-Computer Interaction, particularly with applications in accessibility for blind users. Recently, I built two AI tools, CodeA11y and iTagPDF, that help automatically produce and preserve accessibility metadata for websites and PDFs, respectively.

Accessibility of interfaces through computational methods.

My research lies at the intersection of ubiquitous computing, mobile sensing, human-centered computing, and responsible AI. My work advances privacy-aware sensing systems for health and wellness.

Economics, Consumer Behavior, and Market Psychology.

I am a researcher on the MIND team at Apple. I received my CS Ph.D. student from Stony Brook University, working with Shubham Jain at the intersection of Ubiquitous Sensing and Health Systems in the PiCASSo lab. I have worked on around-the-ear devices. I spent the summer of 2025 with Microsoft Research, Redmond, in the Networking Research Group. My research has been awarded by Stony Brook University and covered by The Economic Times, Outlook, Financial Express, and The Hindu.

I study and design AI and AR interfaces to explore how computational tools can become more transparent, malleable, and dynamic, as well as how they relate to the broader environments they inhabit, primarily in the contexts of accessibility, creativity support, and sensemaking. I believe in empowering people to engage with and through technology in more embodied, pluralistic, and liberatory ways.

I am interested in how AI can enable new, accessible forms of spatial design and artistic expression. As a practitioner in film and design, I am particularly interested in creative agency, including the ability to shape an experience as well as participate in it.

My research explores how AI-enabled personal devices, such as smartphones and wearables, can provide individualized, situated accessibility support that strengthens disabled people’s agency and autonomy. This directly connects to the workshop’s focus on accessibility-centered cyber-physical systems that sense real-world contexts and support everyday activities. Recently, I have been thinking about these goals in terms of end-user programming (primarily natural language programming, but more recently also Programming by Example and Programming by Demonstration). This has led to a project, ProgramAT, which advances this agenda by enabling blind and low-vision people to create and refine bespoke, camera-based, assistive tools, shifting from systems that simply try to adapt to users toward systems users can actively shape.

My research lies at the intersection of AI and HCI, exploring how interactive systems can understand people, context, and the physical world to make interaction more natural and accessible. My recent work focuses on context-aware gesture understanding and vision-based tactile sensing, and I am interested in extending this toward wearable and cyber-physical systems that can adaptively guide, constrain, or augment human movement to support and extend human abilities.

I study accessibility technologies for deaf–hearing communication. We co-designed SignWarriors, a Japanese Sign Language learning game where hearing parents of deaf children summon warriors by signing to a camera. This use of camera sensing to support accessibility-oriented learning aligns with the workshop’s themes, which is why I am joining.

My research explores how XR and AI can augment human cognition. My recent work investigates diminished reality to help students with ADHD manage visual distractions as well as the diminishment and abstract representation of others in shared spaces to support attention while preserving social awareness. More broadly, I’m interested in making everyday environments more accessible and responsive to different cognitive needs.

My research focuses on building interactive audio AI systems to help people make sense of and shape their physical and digital environments as their needs evolve.

My research interests include accessibility-driven design and how we can build interactive systems to support people in their daily lives across a variety of tasks. I'm interested in making complex technologies and information more accessible and exploring how adaptive interfaces can accommodate diverse abilities and preferences, including through multimodal interaction and interdisciplinary approaches, to foster greater independence and agency.

My research interests lie at the intersection of HCI, human perception, and cognitive science. My previous work has involved XR, haptic interfaces, and behavioral experiments, exploring how interactive technologies influence human perception and decision-making. Through these experiences, I became increasingly interested in understanding why people perceive and interpret the same information differently, particularly under uncertainty. Recently, I have been exploring computational approaches to cognition, especially how behavioral experiments and probabilistic models can help explain human perception, judgment, and confidence. Ultimately, I hope to connect these insights with interactive technologies to develop systems that augment human perceptual and cognitive capabilities.

I’m broadly interested in human–AI interaction and accessibility. My research explores how personal AI can understand and expand people’s capabilities and help them develop new ones.

My research explores how sensing and embodied interaction can expand the ways people use everyday systems. In my UIST 2026 paper, I developed a software approach that uses microphone arrays to detect coarse airflow direction, motivated in part by hands-busy situations. I also work on movement-based interaction in XR. These projects connect with the workshop’s focus on context sensing and ability augmentation. I hope to learn from accessibility researchers and discuss how to design and evaluate such systems with diverse users before drawing conclusions about their real-world benefits.