Summary
Kirill Shmilovich is a machine learning scientist with eight years of experience applying deep learning, active learning, and Bayesian optimization to molecular design, computational chemistry, and molecular dynamics. Currently at Genentech after a PhD at the University of Chicago, he has a track record of translating ML-driven discovery into experimentally relevant candidates for π-conjugated peptides and membrane-active small molecules. His work spans adaptive sampling, enhanced-sampling free-energy calculations, and generative backmapping from coarse-grained to atomistic trajectories, bridging physics-based simulation and data-driven models. Previous internships at insitro and Bosch BCAI produced methods that combine docking with ML and neural prediction of molecular wavefunctions, showing his comfort across both algorithmic innovation and domain-specific modeling. Based in Chicago with a strong quantitative background in math, physics, and chemistry, he often tackles problems where physical insight informs model architecture and experimental design.
7 years of coding experience
7 years of employment as a software developer
Bachelor of Science - BS, Mathematics and Physics (double major); Chemistry (minor), Bachelor of Science - BS, Mathematics and Physics (double major); Chemistry (minor) at University of Wisconsin-Milwaukee
Doctor of Philosophy - PhD, Molecular Engineering, Doctor of Philosophy - PhD, Molecular Engineering at University of Chicago
English, Russian