Our CSCW 2026 paper on Situated Practice Systems to support regulation skill development for innovation work received a 🏅 Best Paper Honorable Mention!
Hi, I'm Kapil.
I'm a postdoctoral scholar at the University of California, Irvine, advised by Anne Marie Piper and Erik Sudderth. I develop vision-language model (VLM)-based systems that support visual understanding for blind and low-vision (BLV) people. My research combines human-centered system design and evaluation with ML method development to make AI-powered visual assistance more trustworthy and interactive.

I completed my Ph.D. in Technology and Social Behavior (TSB) (a joint degree in Computer Science and Communication Studies) at Northwestern University, advised by Haoqi Zhang and Darren Gergle. My dissertation argued that developing a capacity for innovation work requires helping students develop their regulation skills, not just domain- or job-specific skills. I developed Situated Practice Systems that use Practice Objects—computational representations of a student's evolving practices and regulation gaps—to support coaching and help students adopt more effective practices, enabling more practice-centric approaches to learning and work.
My research draws from Human-Computer Interaction (HCI), Social Computing, the Learning Sciences, Management and Organizational Sciences, and Artificial Intelligence. My expertise spans HCI systems research—including design research, full-stack system implementation, and mixed-methods evaluation—and applied ML, including dataset annotation, synthetic data generation, foundation model post-training, and human-centered model evaluation.
News
Presented our CHI 2026 paper on image quality issues and VLM captioning at IIT Bombay's IDC School of Design Post-CHI 2026.
Invited to the 10th Annual UCI Postdoc Symposium to give a talk on AI systems for visual accessibility.
Our paper on Situated Practice Systems was accepted to CSCW 2026! 🎉
Our CHI 2026 paper on image quality issues and VLM captioning received a 🏅 Best Paper Honorable Mention!
Our paper on understanding image quality's effect on product captioning with VLMs was accepted to CHI 2026!
Our paper on what helps or hinders remote work at scale was accepted to CSCW 2025! 🎉 This work was conducted during my research internship at Slack, hosted by Mike Massimi.
Publications
Conference & Journal Publications
Situated Practice Systems: A Computational System for Supporting the Coaching and Practice of Regulation Skills for Innovation Work
Kapil Garg, Darren Gergle, and Haoqi Zhang
"It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models
Kapil Garg, Xinru Tang, Jimin Heo, Dwayne R. Morgan, Darren Gergle, Erik B. Sudderth, and Anne Marie Piper
What Remotely Matters? Understanding Individual, Team, and Organizational Factors in Remote Work at Scale
Kapil Garg, Diego Gómez-Zará, Elizabeth Gerber, Darren Gergle, Noshir Contractor, and Michael Massimi
Orchestration Scripts: A System for Encoding an Organization's Ways of Working to Support Situated Work
Kapil Garg, Darren Gergle, and Haoqi Zhang
Understanding the Practices and Challenges of Networked Orchestration in Research Communities of Practice
Kapil Garg, Darren Gergle, and Haoqi Zhang
Opportunistic Collective Experiences: Identifying Shared Situations and Structuring Shared Activities at Distance
Ryan Louie, Kapil Garg, Jennie Werner, Allison Sun, Darren Gergle, and Haoqi Zhang
4X: A Hybrid Approach for Scaffolding Data Collection and Interest in Low-Effort Participatory Sensing
Kapil Garg, Yongsung Kim, Darren Gergle, and Haoqi Zhang