Snehasish Kumar
Staff Software Engineer at Google
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSnehasish Kumar is a Staff Software Engineer at Google with 12 years of experience applying research-grade systems knowledge to make software and hardware run faster. He holds a PhD in Computer Architecture and has deep expertise in cache systems, coherence protocols, workload characterization, and hardware accelerators, work that spans academic publications and industry impact. At Google he focuses on profile-guided optimizations and performance engineering, and his open-source contributions include improving cache simulation in DynamoRIO and adding lifetime profiling to tcmalloc. He also contributed targeted compiler optimizations to LLVM, reflecting a rare blend of low-level systems, memory profiling, and toolchain work. Based in Sunnyvale, he combines rigorous research training with pragmatic engineering to drive measurable speedups in production systems.
12 years of coding experience
10 years of employment as a software developer
B. Tech, Computer Engineering, 8.3/10, B. Tech, Computer Engineering, 8.3/10 at Biju Patnaik University of TechnologySt. James' SchoolDoctor of Philosophy (PhD), Computer Architecture, 4.08, Doctor of Philosophy (PhD), Computer Architecture, 4.08 at Simon Fraser University
Hindi, English
Github Skills (22)
caching10
c-language10
llvm10
testing10
simulations10
memory-management10
instrumentation10
performance-analysis10
profiling10
compiler-optimization10
cprogramming-language10
debug9
unit-testing9
debugging9
binarydiff9
Programming languages (10)
Github contributions (5)
llvm/llvm-project
May 2020 - Apr 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:482 reviews, 16 PRs, 90 pushes in 4 years 11 months
Contributions summary:Snehasish primarily focused on enhancing the Target Library Information (TLI) within the LLVM project. They added support for detecting and optimizing `__size_returning_new` variants, which provide memory allocation size feedback. This involved modifying the TLI definition files, unit tests, and the `SimplifyLibCalls` pass. The user also made improvements to the detection of these library functions, and removed unused code.
compilerstechnologiesclangsubmittoolchain
DynamoRIO/dynamorio
Dec 2017 - Jul 2020
Dynamic Instrumentation Tool Platform
Role in this project:Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:37 commits, 23 PRs, 13 pushes in 2 years 7 months
Contributions summary:Snehasish primarily contributed to the drcachesim tool, enhancing its capabilities for cache simulation and analysis. Their work involved adding functionality to control cache warmup behavior using fractions of loaded blocks, which improved simulation accuracy. They also implemented changes to record and print cache statistics during the warmup phase. Furthermore, they introduced new tests to validate the functionality of the tool.
instrumentationbinary-analysisprofilingwindowslinux
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.