Joseph McMahan, Ph.D. is a computer architecture researcher focused primarily on how architecture intersects the worlds of deep learning, security, and formal methods. He publishes in top computer architecture conferences (ISCA and ASPLOS) and has twice had a paper selected for an IEEE Micro Top Pick. His PhD thesis was awarded honorable mention for the 2020 ACM SIGARCH/IEEE CS TCCA Outstanding Dissertation Award.
Joseph's work is motivated by the question "How do we build better systems?" Looking beyond tradtional metrics for "better," there are a wealth of ways in which systems can be studied and improved --- metrics like security, analyzability, and designer effort often fall by the wayside, or are considered only as an afterthought. Frequently, in pursuit of these metrics, we can leverage techniques from the programming languages community to get better analyses and design methodologies. Joseph's research has spanned novel architectures to improve software reasoning and verification, new methods of quantifying hardware side-channel leakage, and tools to improve automation in deep learning accelerator design.
After receiving his B.A. in Physics from Princeton University, Joseph worked under Timothy Sherwood at UC Santa Barbara for his Ph.D. in computer architecture. He is currently employed as a research scientist at the University of Washington's Paul G. Allen School of Computer Science & Engineering in the SAMPL group with Luis Ceze.
In his free time, Joseph enjoys video games, fiction, poetry, programming, and music. His favorite things are fresh-roasted coffee and walks with Harper (dog) and Jonathan (human).
Joseph is actively seeking employment opportunities beginning in summer 2021. Feel free to take a look at a hopefully recent copy of his C.V. You can reach him via the email at the bottom of this website.