Summary
Walker Willetts is a Senior Architect specializing in cycle-accurate GPU and SoC performance modeling with a decade of experience exploring microarchitectural tradeoffs for ARM CPUs and GPUs. Now at NVIDIA after leading CPU performance modeling at Arm and Microsoft, he writes production-grade Python and C++ frameworks that connect simulation, RTL correlation, and data visualization to guide silicon decisions. His work spans L2/SLC cache modeling, CHI protocol features, execution-unit design space exploration, and tooling that speeds simulation warmup and profiling. Trained as an electrical engineer with an M.Eng in Quantum Computing from Duke, he combines deep math and physics instincts with practical software craftsmanship. Notably, he has built bridges between software models and SystemVerilog for unit-level correlation—a recurring theme that blends verification with performance insight.
10 years of coding experience
5 years of employment as a software developer
Master of Engineering - MEng Computer Engineering - Quantum Computing, Master of Engineering - MEng Computer Engineering - Quantum Computing at Duke University
High School Diploma STEM Track, High School Diploma STEM Track at West Forsyth High School
No degree earned - courses taken while in high school Organic Chemistry, No degree earned - courses taken while in high school Organic Chemistry at Wake Forest University
English