Mitchell Koch (BS ’11) is a Manager and Senior Data Scientist for Machine Learning at DoorDash. He was part of the initial team focusing on the core logistics problem including time prediction and vehicle routing and now leads machine learning initiatives including for personalized recommendations and marketing optimization.
He worked on Devika Subramanian’s research team the entire time he was at Rice, focusing on the application of Bayesian network learning methods to the T-cell signaling network and to differential equations models for validation. He wrote Bayesian network learning software using dynamic programming methods, and applied texture analysis features and supervised learning to the problem of identifying four types of mucosal patterns in Barrett’s Esophagus.
His research interest shifted at the University of Washington, where he completed his M.S. in 2014. There, he focused on problems in natural language processing and information extraction.