Yeonghun Kang is a Vector Institute researcher and a postdoctoral researcher in the Matter Lab at the University of Toronto, working under the supervision of Prof. Alán Aspuru-Guzik. His research lies at the intersection of materials science and artificial intelligence, focusing on advancing machine-learning methods for porous materials. Trained as a computational materials scientist during his PhD at KAIST with Prof. Jihan Kim, he has developed multi-modal predictors, generative models, and autonomous AI systems that link structure–property understanding with rational materials design. His work integrates molecular simulations, density functional theory, and high-throughput computation with representation learning to reveal chemical patterns and accelerate materials discovery.

Kang’s recent research extends into agentic AI for chemistry, building large language model–driven agents that can reason about experiments, ensure laboratory safety, plan synthesis strategies, and collaborate with self-driving lab platforms. He also develops large-scale scientific data-mining pipelines that extract synthesis conditions, properties, and chemical knowledge from literature and images, enabling high-quality datasets for downstream modeling. Across his projects, Kang aims to create a unified framework in which material prediction, synthesis planning, autonomous experimentation, and knowledge extraction operate seamlessly. His long-term vision is to build AI systems that not only model chemical reality but also autonomously act within it—accelerating, systematizing, and democratizing materials innovation.