Vision Foundation Models · LLM Post-Training · Medical AI
I'm an AI Research Scientist at Lunit working on foundation models and post-training for medical AI.
I'm interested in how learned systems behave under change: how they adapt, generalize, forget, exploit objectives, and fail when deployed beyond their training distribution. My work spans representation learning, post-training, evaluation, and increasingly agentic systems, with a focus on understanding and improving model reliability.
Previously, I worked on pathology foundation models at Lunit, Cell Painting foundation models at KAIST, and computer vision for drug discovery at Institut Pasteur.