Summary
Sangwon Jeong is a PhD student in Computational Science at Vanderbilt University with nine years of experience applying generative models and explainable AI to real-world problems. He focuses on uncovering and manipulating semantic concepts in latent spaces of VAEs, GANs, and diffusion models, combining interactive visualization and user feedback to make these models more accessible to both experts and casual users. His research spans image synthesis, LLM evaluation and intervention, and domain-specific languages for model control, with publications in top visualization venues like EuroVis and IEEEVis. Sangwon translated academic work into applied impact during an internship at Lawrence Livermore National Laboratory, where he built a visualization tool to browse 80,000+ AI-generated images for materials discovery. Comfortable bridging HCI and ML, he brings hands-on engineering from co-founding a startup to deploying cloud-backed mobile apps and production ML tools. He is driven to make multi-modal latent spaces intuitive for human users, enabling finer-grained manipulation and faster scientific discovery.
9 years of coding experience
Bachelor's degree, Business Administration and Management, Finance, 3.44, Bachelor's degree, Business Administration and Management, Finance, 3.44 at Seokyeong University
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Vanderbilt University