Profile
I am a first-year M.S. student in Mobility at the Cho Chun Shik Graduate School of Mobility, Korea Advanced Institute of Science and Technology (KAIST), advised by Prof. In-Hi Kim. I previously studied Urban Planning & Engineering at Yonsei University with a minor in Applied Statistics.
My goal is to build safe foundation-AI systems for real transportation decisions. I am especially interested in cases where conventional models have little data to learn from: new mobility services, expanding choice sets, heterogeneous traveler responses, and complex vehicle–infrastructure interactions.
- Multi-agent LLMs
- Travel behavior
- Discrete choice
- Physical AI
- Autonomous driving
- Mobility data
Research path
KAIST · 2026–present
Developing city-scale LLM-agent simulations of human travel behavior and cooperative world models for autonomous driving.
Yonsei · 2025–2026
Built an LLM-agent RP–SP simulator for individual-level demand prediction of autonomous shuttles and demand-responsive transit.
StellarVision · 2024–2025
Worked on satellite-image object detection and spatial analysis for maritime-logistics applications.
How I work
Behavior before benchmark
I start with the decision process and human context, then ask what a model's score actually represents.
Evidence before claim
I separate promising demonstrations from validated findings and keep evaluation tied to held-out human observations.
Systems, not isolated models
I combine agents, retrieval, behavioral constraints, simulation, and domain tools into end-to-end research workflows.
Reproducible communication
I aim to make research inspectable through clear methods, code, visualizations, and paper materials.