D.E. Shaw Research (DESRES) develops and uses advanced computational technologies to understand the behavior of biologically and pharmaceutically significant molecules at an atomic level of detail, and to design precisely targeted, highly selective drugs for the treatment of various diseases. Among its core technologies is Anton, a proprietary special-purpose supercomputer that DESRES designed and constructed to vastly accelerate the process of molecular dynamics simulation.

Following the talks, guests are invited to stay for an informal get-together with snacks and drinks, where they will have the opportunity to learn more about working at DESRES and connect with members of the team.

About the Speakers

Dr. Leon Klein

Physics alumnus of Freie Universität Berlin, Dr. Leon Klein earned his PhD in mathematics in Professor Frank Noé’s AI4Science research group. He specialized in machine learning, generative models, and molecular sampling methods, focusing on improving molecular simulations through the integration of machine-learning models. As a visiting researcher at Microsoft Research, Leon Klein led the Timewarp project, which aimed to accelerate molecular simulations in a transferable way using normalizing flows.

Dr. Sebastian Dick

Physics alumnus of the University of Würzburg, Dr. Sebastian Dick earned his PhD in physics at Stony Brook University. Sebastian holds a Ph.D. and an M.A. in physics from Stony Brook University, and a B.Sc. in physics from the University of Würzburg in Germany.  Most recently, Sebastian was a senior investigator on the computational molecular physics team at Roivant Discovery in New York, where he developed, implemented, and validated molecular mechanics charge models based on deep neural networks.  As a graduate research assistant in Marivi Fernandez-Serra’s group at Stony Brook University, Sebastian investigated how machine learning and density functional theory can work together to achieve improved electronic-structure calculations, and he developed NeuralXC, a user-friendly Python framework, to train and utilize deep learning models for electronic structure calculations.