Research¶
Kevin Ryczko is a researcher and machine-learning scientist. He obtained a Ph.D. in Physics from University of Ottawa (in collaboration with the National Research Council of Canada and the Vector Institute for Artificial Intelligence). Before that he obtained his Masters degree in modelling and computational science and Bachelor degree in Physiscs at the University of Ontario Institute of Technology (UOIT, now Ontario Tech University).
🔬 Research background & contributions¶
Early in his academic career, he conducted research on liquid water-metal interfaces, using ab initio electronic structure methods (density functional theory, first-principles molecular dynamics) to study how water behaves at metal surfaces — relevant for understanding water electrolysis and nanoscale energy/chemistry phenomena. As an undergrad, he even built a simulation package called “hashkat” to study how online social networks evolve, how connections form, and how information propagates.
Kevin Ryczko has also contributed to the field of deep learning for physical sciences. During his PhD, he developed machine learning models to predict properties of molecular systems, such as molecular dipole moments and potential energy surfaces. His work demonstrated how ML can accelerate simulations while maintaining accuracy, enabling the study of larger systems and longer timescales.
🧑💻 Current role & work in AI / Materials / Drug Design¶
Today, Kevin Ryczko works at SandboxAQ, where he applies deep learning / machine learning techniques to problems in computational chemistry, materials science, and biology. For instance, he is among the authors of a paper describing IDOLpro — a generative-AI platform for structure-based drug design. This AI combines diffusion models with multi-objective optimization to generate novel chemical compounds (ligands) with favorable properties like binding affinity and synthetic accessibility — potentially speeding up drug discovery. According to his work profile, his focus includes using ML to accelerate computation and design of nanoscale materials.
🎯 Significance¶
Kevin Ryczko represents a growing class of scientists bridging physics / quantum-chemistry + machine learning / generative AI. His trajectory — from studying fundamental physical interfaces to working on AI-driven material/chemical discovery — highlights how modern research increasingly integrates traditional scientific modelling with cutting-edge AI.