Pate Motter
Software Engineer - TPU Performance at Google
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolPate Motter is a performance-focused software engineer with a PhD in computer science and 11 years of experience optimizing HPC and ML workloads across industry and research. Currently at Google, he works on inference performance for TPUs after improving latency for generative models on GPUs and AWS Inferentia at Amazon. His background spans low-level numerical linear algebra, heterogeneous computing (OpenCL, MPI, OpenMP), and production ML systems, reflecting a knack for squeezing out performance across hardware stacks. Notably, his doctoral work applied machine learning to select solver-preconditioner pairs tuned to available hardware—an approach he continues to bring to production inference optimization. Based in Seattle, he combines rigorous research training with hands-on engineering to make code go fast in real-world systems.
11 years of coding experience
7 years of employment as a software developer
Master of Science (M.S.), Computer Science, Master of Science (M.S.), Computer Science at University of Colorado at BoulderB.S., Computer Science, B.S., Computer Science at University of Arkansas at Fayetteville
Stackoverflow
1,043reputation
144kreached
2answers
7questions
jupyter
top-5%
pycharm
top-5%
Github Skills (66)
gpt10
large-language-models10
llm10
benchmark9
gpu9
pytorch9
inference9
pycharm9
jupyter9
transformer8
hpc-applications8
llm-inference8
gemma8
mlops8
mnist8
Programming languages (5)
Github contributions (5)
patemotter/trilinos-prediction
Aug 2016 - Aug 2017
Contributions:212 pushes, 3 branches in 1 year
LighthouseHPC/trilinos-petsc-testbed
Dec 2015 - Feb 2017
Contributions:62 pushes, 1 branch in 1 year 2 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.