Hello! I'm an undergraduate student at KAIST studying STS and Mathematical Sciences, with a minor in Design. My work sits at the intersection of social AI, multi-agent systems, and human-computer interaction. I am especially interested in how AI systems can understand people and groups—and how they can augment human creativity, collaboration, and decision-making.
I currently conduct research with Asuman Özdağlar at MIT LIDS, with Tak Yeon Lee at AI Experience Lab. Across these collaborations, I combine computational methods, system building, and human-centered evaluation to study how intelligent systems behave within real social and organizational contexts. Outside of research, I secretly pursue theater and stand-up comedy. Check out my creative page to see this side of me.
I also keep a research manifesto on the research I want to pursue, the problems I hope to solve, and the kind of life I want to live.
TIDES: A Longitudinal Bilingual Dataset for Modeling Multi-Party Social Dynamics.
COLM · 2026
Heechan Lee, Jeonggyu Kang, Junho Myung, Jaywoong Jeong, Juho Kim, Joseph Seering
Gaze2Prompt: Turning Eye-Tracking Data into Visual Prompts for Multimodal LLMs.
ACM UbiComp/ISWC Poster · 🏆 Best Poster Award · 2025
Jae Young Choi, Seon Gyeom Kim, Jaywoong Jeong, Ryan Rossi, Jihyung Kil, Tak Yeon Lee
TeamGym (working title).
Preprint · Under review · 2026
Chanwoo Park, …, Jaywoong Jeong, …, Asuman Özdağlar, et al.
Research Assistant
Mentor: Chanwoo Park|Advisor: Prof. Asuman Özdağlar
Co-authoring TeamGym and studying how strategic LLM agents behave in multi-level organizational structures under human-inspired constraints.
Research Intern
Mentors: Jae Young Choi, Seon Gyeom Kim|Advisor: Prof. Tak Yeon Lee
Led DramaForge and co-authored Gaze2Prompt, developing AI systems for constraint-aware creative work and gaze-grounded multimodal interaction.
Research Assistant
Mentor: Heechan Lee|Advisor: Prof. Joseph Seering
Co-authored TIDES and built data-processing and annotation workflows for longitudinal, bilingual, multi-party collaboration data.
Research Assistant
Advisor: Prof. Johan Chu
Studied shifts in film-industry participation after Netflix and generative AI, and built an LLM classification pipeline over 112M news articles.
MIT Media Lab · Center for Constructive Communication · With Brandon Roy
Mapping shared themes, disagreements, and perspective structures while evaluating whether LLM-generated completions preserve the diversity of human contributions.
University of Utah · Carnegie Mellon University · With Yue Jiang, Jeffrey P. Bigham, Jae Young Choi
Aligning eye-tracking traces with interactive web components and designing WebArena tasks for gaze-informed web access for blind and low-vision users.
MIT Sloan School of Management · With Robin Na, Michiel Bakker
Extending an initial workshop study into auditable simulations of aggregate electoral opinion trajectories under evolving media exposure.
KAIST ID301 Interactive Product Design Term Project. Turning abstract time into a sensory, intuitive experience with a physical–software hybrid timer.
Poster design for a theatre production.
Built RAG-based financial information retrieval and QA system. Proposed Query Chain framework for latent user intent analysis. 4th UNIST-KAIST-POSTECH Data Science Competition 3rd place.
Feel free to or find a time on Calendly.