Summary
Raunak Kumar is an applied scientist at Microsoft Turing working on post-training reasoning models for Microsoft 365 Copilot, with a decade of experience bridging research and production ML systems. He earned a PhD in computer science from Cornell (advised by Kleinberg and Dean) after a BSc at UBC, and his thesis explored online decision-making and network-science perspectives on learning-from-interactions. Raunak’s work spans online learning, bandits, optimization, control, and practical graph-structured ML—he contributed to ranking and search features in Copilot and scaled Graphormer for link prediction in Office 365. A repeat Microsoft intern who turned research insights into deployable solutions, he’s equally comfortable proving convergence rates (EM) as he is prompting LLMs to translate queries into graph queries. He has a strong teaching pedigree with multiple TA awards and a knack for turning theoretical ideas into robust engineering outcomes.
10 years of coding experience
6 years of employment as a software developer
Bachelor of Science, Computer Science, Bachelor of Science, Computer Science at The University of British Columbia
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Cornell University
Elementary, Middle and High School, Elementary, Middle and High School at Birla High School
English, Hindi, Bengali, Maithili, Spanish, French