Saumil Patel
Computational Scientist - Computational Science Division & Leadership Computing Facility
Greater Chicago Area United States
Summary
👤
Senior🎓
Top SchoolSaumil Patel is a computational scientist with over a decade of experience optimizing and integrating multi-physics scientific applications for DOE leadership-class supercomputers. At Argonne he develops high-performance kernels (CUDA, OpenMP, DPC++/SYCL, OpenACC), co-leads DOE-funded projects for power-generation simulations, and coaches domain scientists on profiling, parallelization, I/O, and workflow design to reduce time-to-solution. His PhD work produced NekLBM, a high-order CFD solver used for moving-boundary simulations, and his postdoctoral optimizations sped multi-physics runs by roughly 3x on leadership systems like Theta. Pragmatic about peak occupancy and FLOPs, he combines deep numerical-methods knowledge with hands-on tuning across Intel, AMD, and NVIDIA architectures. He was part of an HPCwire-recognized team for exemplary HPC use in energy applications, and he enjoys the continual puzzle of squeezing more performance from evolving hardware.
9 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy - PhD, Mechanical Engineering, Doctor of Philosophy - PhD, Mechanical Engineering at The City College of New YorkMaster of Arts (M.A.), Applied Mathematics, Master of Arts (M.A.), Applied Mathematics at Hunter CollegeBachelor of Arts (B.A.), Mathematics and Economics, Bachelor of Arts (B.A.), Mathematics and Economics at Cornell University
Gujarati
Github Skills (11)
sem10
spectral10
fluid8
hpc5
openfoam4
chemical-engineering3
hydrodynamics3
computational-fluid-dynamics3
testing3
java3
modeling-tool3
Programming languages (2)
Github contributions (5)
spatel7286/hello-world
Dec 2017 - Jul 2021
Contributions:2 PRs, 3 pushes, 2 branches in 3 years 7 months
testingtest-repository
spatel7286/Nek5000
Aug 2016 - Dec 2016
Contributions:1 push, 1 branch in 3 months
scalablespectralportablecfd
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.