Christopher Kelly is a Research Staff 4 and Applications Team Lead with over 15 years of high-performance computing and computational physics experience, driving scalable scientific software and algorithms on DOE leadership-class supercomputers. He leads multidisciplinary teams that deliver performance-optimized solutions across AI/ML, lattice QCD, Bayesian UQ, and real-time performance monitoring, and has taken projects from algorithm design to 10,000-GPU scaling runs and 300 PFLOP demonstrations. His contributions include pioneering lattice QCD calculations of kaon direct CP violation, inventing "X-conjugate" algorithms that cut ensemble and solver time dramatically, and authoring award-winning publications while single-handedly delivering software on Frontier, Aurora, Perlmutter, and Polaris. He also migrated large transformer models to DeepSpeed for fusion disruption prediction and built Chimbuko, a real-time anomaly detector that scales to thousands of nodes. Based in the Village of Delhi, NY, he blends deep theoretical physics training (PhD, Edinburgh) with hands-on leadership in codesign, mixed-precision optimization, and Bayesian modeling to shorten scientific timelines by years.
10 years of coding experience
4 years of employment as a software developer
Ph.D., Theoretical Physics, Ph.D., Theoretical Physics at The University of Edinburgh
MPhys, Physics, 1st with Honours, MPhys, Physics, 1st with Honours at University of Oxford
Contributions:72 pushes, 21 branches in 9 years 3 months
cppparallelc-plus-plusparallel-datamathematical
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.