Devangi Parikh
Associate Professor Of Instruction
Austin, Texas, United States
Summary
🤩
Rockstar🎓
Top SchoolDevangi Parikh is an Associate Professor of Instruction in Computer Science at UT Austin with eight years of professional experience and a PhD from Georgia Tech. She blends academic teaching and mentorship with deep low-level systems expertise, having contributed assembly-optimized kernels and AVX-512 performance tuning to the BLIS high-performance linear algebra framework. Based in Austin, she brings practical experience in hardware-specific optimization—tuning prefetching, instruction selection, and register usage for SkylakeX—to courses and projects that bridge theory and real-world performance. Her profile reflects a rare combination of pedagogy and hands-on performance engineering, making her especially effective at training the next generation of systems programmers.
8 years of coding experience
Doctor of Philosophy (PhD), Doctor of Philosophy (PhD) at Georgia Institute of TechnologyBachelor of Engineering (BE), Bachelor of Engineering (BE) at Nirma Institute of Technology
Github Skills (14)
avx10
matrix-multiplication10
cluster-computing10
assembly10
parallel-computing10
blas10
performance-optimization10
x86-6410
scientific-computing10
linear-algebra10
assembler10
optimization9
c119
c179
Programming languages (3)
Github contributions (5)
flame/blis
Dec 2017 - Feb 2019
BLAS-like Library Instantiation Software Framework
Role in this project:
Performance Engineer & Assembly Programmer
Performance Engineer & Assembly ProgrammerContributions:18 commits, 12 pushes, 1 branch in 1 year 2 months
Contributions summary:Devangi's contributions focus on low-level performance optimization within the BLIS (BLAS-like Library Instantiation Software Framework) repository. They primarily added and modified assembly kernels, specifically for SkylakeX (SKX) and potentially other architectures, aimed at improving the performance of matrix multiplication (dgemm, sgemm). Their work involves careful tuning of prefetching, instruction selection, and register usage to achieve optimal performance, with a specific focus on AVX-512 instructions. Additionally, the user contributed to hardware-specific test drivers.
software-frameworklinear-algebra-librarylinear-algebramatrix-multiplicationblis
dnparikh/server
Aug 2018 - Aug 2018
Server for gathering and displaying OER metadata and topic tags
Contributions:4 PRs, 38 pushes in 2 days
gatheringoermetadatatags
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.