Gábor Buella
Software Engineer at Google
Warsaw, Masovian Voivodeship, Poland
Summary
🤩
RockstarGábor Buella is a seasoned software engineer with 12 years of experience building high-performance backend and systems software across industry leaders including Google, Oracle and Intel. He has deep expertise in low-level systems programming, performance optimization and CPU-specific tuning, demonstrated by contributions to high-profile open-source projects like PMDK, oneDNN and Clang. His work spans persistent memory primitives, deep learning library optimizations and compiler intrinsics, showing a rare blend of practical benchmarking, hardware-aware codegen and test-infrastructure improvements. Based in Warsaw, he brings production experience from both large enterprises and cross-border teams, having worked in multiple countries and contexts. Colleagues rely on him for pragmatic fixes that improve correctness and measurable throughput rather than flashy rewrites. He pairs a systems mindset with attention to test coverage and reproducible benchmarks, making complex low-level changes safe to ship.
12 years of coding experience
17 years of employment as a software developer
Hungarian, English
Stackoverflow
1,880reputation
65kreached
49answers
19questions
Github Skills (38)
algorithm10
code-optimization10
algorithms10
c-language10
intrinsics10
performance-monitor10
edn10
system-programming10
avx10
low-level10
c1110
x8610
c1710
low-level-programming10
compiler-design10
Programming languages (8)
Github contributions (5)
pmem/pmdk
Aug 2016 - Nov 2017
Persistent Memory Development Kit
Role in this project:
Back-end Developer
Back-end DeveloperContributions:54 commits, 59 PRs, 226 comments in 1 year 3 months
Contributions summary:Gábor primarily focused on low-level systems programming and performance optimization within the Persistent Memory Development Kit (PMDK). They implemented default settings for non-temporal store (movnt) memory copies and addressed a format string issue. The user also contributed to fixing a calculation in a benchmark and fixing isblank usage. Other contributions include renaming a confusing macro and providing test improvements to include more address data.
pmdkpmemdevelopment-kitmemorypersistent-memory
uxlfoundation/oneDNN
Jun 2020 - Apr 2021
oneAPI Deep Neural Network Library (oneDNN)
Role in this project:Backend Developer
Backend DeveloperContributions:10 commits in 10 months
Contributions summary:Gábor's commits primarily involve modifications and improvements to the oneDNN library's core functionalities, focusing on performance optimizations for specific hardware architectures. They refactored and tuned the preamble section of the x64 CPU implementation for reorder operations, potentially improving efficiency. Furthermore, the user addressed issues related to consistency checks in the pooling module and incorporated scale attributes in binary operations, showcasing involvement in algorithm implementation and performance enhancements. They also contributed to test and benchmark infrastructure.
bfloat16sse41avx512openmpopencl
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.