Kim Hazelwood

Kim Hazelwood is a computer systems researcher and technology executive whose 25-year career has moved between universities, national laboratories and the companies that build the world’s largest computing infrastructure. In October 2025 she was named an at-large member of the board of directors of Empire AI, the New York consortium of public and private research institutions that pools computing capacity for academic researchers. That makes her one of the few people anywhere with a sitting governance seat on a public AI computing consortium — the same class of institution California’s framework work is trying to design.

She began in academia. After a Ph.D. in computer science from Harvard University in 2004 and an undergraduate degree in computer engineering from Clemson University, she joined the University of Virginia, where she became the first woman to earn tenure in its computer science department. Her research there covered optimizing compilers, dynamic binary translation and workload characterization, and produced “Dynamic Binary Modification: Tools, Techniques, and Applications,” published in 2011. She was a co-author of Pin, the dynamic binary instrumentation framework that became standard equipment for a generation of computer architecture researchers, and the work earned the ACM SIGPLAN 10-Year Test of Time Award in 2015.

Her industry career tracks the build-out of modern AI infrastructure. She was a research scientist at Intel, a software engineer in Google’s datacenter division — where she contributed to the Tensor Processing Unit — and director of systems research at Yahoo Labs. At Meta she held a series of engineering leadership roles across Infrastructure and Research, rising to senior director of engineering and research, and contributed to the AI Research SuperCluster, one of the largest machine learning training systems built to date. The through-line is unglamorous and directly relevant to a public compute platform: what a fleet of machines actually costs, how it should be sized and what it takes to keep researchers productive on it.

She has published more than 50 peer-reviewed papers and books across compilers, computer architecture and applied machine learning, and has kept up an unusually heavy load of public service for someone in industry. She served on the board of directors of the Computing Research Association from 2017 to 2020 and has advised the National Science Foundation and the National Academies, along with advisory boards at MIT SystemsThatLearn, EPFL EcoCloud and Bruin AI at UCLA. Her awards include the MIT Technology Review Top 35 Innovators Under 35, the CRA-W Anita Borg Early Career Award, an NSF CAREER Award, a NASA New Investigator Award and a place on the 2025 Top 100 Women in AI list.