Piotr Padlewski
Member Of Technical Staff at Anthropic
Zurich, Zurich, Switzerland
Summary
👤
Senior🎓
Top SchoolPiotr Padlewski is a backend-focused software engineer with a decade of experience specializing in C++, compiler work and high-performance systems, now based in Zurich. He has contributed directly to LLVM/Clang—implementing devirtualization optimizations, vtable generation improvements, and clang-tidy checks—bringing compiler-level performance wins to real-world code. His recent roles span cutting-edge multimodal ML at Anthropic and Reka AI and long-term work on large vision-language models at DeepMind, combining systems optimization with ML research. Piotr’s background includes low-level runtime tuning on NVIDIA Jetson for autonomous drone racing and compiler intern work at Microsoft and Google that produced measurable ThinLTO and PGO improvements. He blends production-grade engineering with research curiosity, often tackling optimizations few engineers see, such as emitting assumption loads to enable devirtualization.
10 years of coding experience
10 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at University of WarsawI Liceum Ogólnokształcące im. Adama Mickiewicza w Białymstoku
Polish, English, German
Github Skills (14)
static-code-analysis10
code-analysis10
compiler-design10
code-generation10
c-language10
refactoring10
clang-tidy10
cprogramming-language10
llvm9
cplus9
compiler-compiler9
cpp9
compiler9
boost8
Programming languages (12)
Github contributions (5)
llvm-mirror/clang-tools-extra
Apr 2016 - Mar 2017
Mirror kept for legacy. Moved to https://github.com/llvm/llvm-project
Role in this project:
Back-end Developer
Back-end DeveloperContributions:13 commits in 10 months
Contributions summary:Piotr primarily focused on contributing to clang-tools-extra, a collection of clang-based tools. Their contributions involved modifying and adding checks within clang-tidy, a tool for static code analysis and refactoring. The user implemented and enhanced checks related to code style, specifically addressing the use of `boost::lexical_cast` and `push_back` in favor of more modern and efficient alternatives like `std::to_string` and `emplace_back`. They also added and fixed bugs in checks that involve smart pointers.
keptwindowsllvm
llvm-mirror/clang
Jul 2015 - Jul 2018
Mirror kept for legacy. Moved to https://github.com/llvm/llvm-project
Role in this project:Back-end Developer
Back-end DeveloperContributions:45 commits in 2 years 11 months
Contributions summary:Piotr primarily focused on modifying the Clang compiler's code generation process, specifically related to virtual function handling and vtable generation for C++ code. Their work included implementing optimizations, fixing bugs, and introducing features like emitting assumption loads for vtable pointers to aid in devirtualization, which can improve performance. They also introduced mechanisms for handling virtual functions with respect to inline functions, and contributed to the logic for emitting vtables opportunistically.
keptwindowsllvmcc-plus-plus
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.