Matthew Rosoff

Matthew Rosoff is a principal data scientist in Microsoft’s Applied Sciences Group (ASG), where he leads teams advancing generative A.I. capabilities for Windows.  His work focusses on customer-representative synthetic data generation for model training, model fine-tuning, and responsible A.I. testing.  He has research interests in synthetic data generation, differentially private data generation, scalable database systems, and A.I.-based database/analytics systems.

Before joining ASG, Matthew was a principal data scientist lead for Azure Cloud, where he managed teams improving the efficiency, performance, and quality of service for Microsoft’s largest data storage and database technologies, including Cosmos and Cosmos DB.  His leadership resulted in over $100 million in annual savings through innovations such as machine-learning-powered capacity reservation, resource prediction, and query optimization systems.  Earlier, as a data scientist for Windows, he solved engineering challenges in build efficiency, release signoff, and browser performance using advanced M.L. and statistical methods.  He also served as a program manager for SQL Azure, building analytics platforms and improving service quality through big data analysis.

Matthew holds both a master’s and a bachelor’s degree in electrical and computer engineering from Cornell University.  He also earned a professional certificate in data science from the University of Washington.

Outside of work, Matthew enjoys skiing, fishing, biking, and hiking with his wife and dog.