Research
Research Interests
I believe the most impactful AI research happens at the boundary between technology and human experience. My work is driven by a simple question: how do we build intelligent systems that genuinely help people - not just in controlled lab settings, but in the messy, unpredictable flow of everyday life?
Focus Areas
Human-Centered AI
Designing AI systems that prioritize human values, capabilities, and experiences. My work focuses on how AI can augment human intelligence while preserving agency, interpretability, and meaningful control - especially in complex domains like healthcare, learning, and decision-making.
LLM Applications
Building practical applications powered by large language models that seamlessly integrate into real-world workflows. This includes conversational agents, RAG-based systems, and multi-agent pipelines that enhance productivity, reasoning, and information access.
AI-Assisted Learning
Exploring how AI can support self-directed learning through adaptive feedback, explanation, and reflection. My work investigates how LLM-based systems can provide principle-based guidance, improve skill acquisition, and reduce dependency on external instruction.
Reliable AI
Developing methods to ensure AI systems are trustworthy, robust, and aligned with user expectations. This includes evaluation frameworks, uncertainty handling, and design strategies that promote transparency, controllability, and safe deployment of AI in high-stakes settings.
Current Position
M.S. Candidate - Human-Centered Intelligent System Lab
GIST (Gwangju Institute of Science and Technology) · Mar. 2025 – Present
Advisor: Prof. SeungJun Kim
Background & Motivation
My journey started in software engineering - building products at a startup, competing in hackathons, and learning what it means to ship real software to real users. That grounding keeps me focused on practicality: research should ultimately be deployable.
Through internships at SNUBH Medical AI Center and GroupByHR, I saw firsthand how AI could either empower or frustrate people depending on how thoughtfully it was designed. Those experiences shaped my conviction that HCI and AI belong together.
Today, I design and study AI systems through the lens of the people who use them - asking not just “does this work?” but “does this help, and why?”