Publication Apr 9, 2026
Paper Accepted at IEEE Transactions on Vehicular Technology
Our paper on FedKuramoto was accepted at IEEE Transactions on Vehicular Technology (JCR IF 7.1).
Welcome! I'm Jungwon Seo
Ph.D. Candidate in Applied Data Science at University of Stavanger
Jungwon Seo is a Ph.D. candidate in Applied Data Science at the University of Stavanger, funded through the NCS2030 project. His work spans academia and industry, with hands-on experience across research, software engineering, and teaching in both South Korea and Norway.
His research focuses on Federated AI: privacy-preserving collaborative AI, including reliable federated training, practical deployment, and secure inference in decentralized environments. More recently, he has been expanding this direction through a privacy-preserving lens, with particular interest in sensitive information leakage risks in AI agents. Alongside research, he has built and operated full-stack systems end-to-end, from frontend and backend development to cloud deployment and production operations.
He received his M.S. in Computer Science from the University of Stavanger, with a master's thesis titled Minimum Word Error Rate Training for Speech Separation, and his B.S. in Computer Science from Yonsei University. He has worked as a founder or founding engineer at multiple startups, interned at global companies including Amazon Web Services and Equinor, and taught Big Data, Machine Learning, and AI courses at Yonsei University Graduate School of Information.
What I Do
Research
Practical research centered on Federated Learning, expanded through a privacy-preserving lens. Current focus: sensitive information leakage risks in AI agents.
Software Engineering
Full-stack engineering from CSS to DevOps, with hands-on startup experience building systems from scratch and running deployment/operations in production.
Teaching
Hands-on teaching in SWE, AI, and ML, with content flexibly tailored for both academic and industry audiences.
News
Ph.D. Apr 8, 2026
Ph.D. Thesis Officially Submitted
Official submission of my Ph.D. thesis, Toward Federated AI: Integrating Reliable Training, Automated Deployment, and Secure Inference.
Publication Mar 3, 2026
Published: Trustworthy AI in Safety-Critical Systems
Publication of work on robustness, adversarial resilience, and uncertainty-aware intelligence for safety-critical AI systems.
Talk Jan 19, 2026
Talk at VÃ¥r Energi Stavanger Office (NCS2030 Project)
Presented Federated Knowledge Cloud for Subsurface Digitalization at VÃ¥r Energi Stavanger Office under the NCS2030 project.
Publication Jan 12, 2026
GC-Fed Accepted at Information Fusion
GC-Fed accepted at Information Fusion (JCR IF 15.5), a top-tier AI journal.
Publication Dec 24, 2025
FedShift Published in IEEE Access
FedShift was published in IEEE Access on December 24, 2025.
Publication Dec 22, 2025
Publication from Master Thesis Supervision (Espen Sele)
Secure federated learning via neural cryptography with homomorphic operations was published from a co-supervised Master thesis project.
Publication Oct 2, 2025
ProxyLLM Published in ECML PKDD 2025 Demo Track
ProxyLLM was accepted and published in the ECML PKDD 2025 Demo Track, co-authored with Sehyeong Jo.