Michał Tarnowski
Performance Engineer at Ampere
Warsaw, Masovian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolMichał Tarnowski is a Performance Engineer based in Warsaw with a decade of experience optimizing AI and graphics workloads for mobile and embedded platforms. He blends computer vision, computer graphics, geometry processing and deep learning expertise, having moved from junior engineering roles to senior AI and performance-focused positions at TCL Research Europe and Ampere. Michał contributes to prominent open-source projects like the Xiaomi MACE inference framework, where he improved reliability and parallel GPU performance through bug fixes, tests and datatype optimizations. Known for pragmatic, low-level optimizations and careful test-driven improvements, he reliably turns research prototypes into production-ready, high-performance implementations.
10 years of coding experience
7 years of employment as a software developer
Bachelor's degree, Bachelor's degree at University of Warsaw
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (10)
opencl10
gpu-programming10
machine-learning10
deeplearning-ai10
deep-learning10
tensorflow9
cprogramming-language9
neural-network9
c-language9
testing8
Programming languages (6)
Github contributions (5)
XiaoMi/mace
Jan 2019 - Feb 2019
MACE is a deep learning inference framework optimized for mobile heterogeneous computing platforms.
Role in this project:
ML Engineer
ML EngineerContributions:7 commits, 4 PRs, 1 comment in 21 days
Contributions summary:Michał contributed to the deep learning inference framework by fixing bugs, adding tests, and optimizing existing code. Their work includes adding pragmas for parallel processing, changing data types, and addressing issues with GPU Eltwise operations. They also added additional tests and addressed NaN propagation issues. These changes show a focus on improving the performance and reliability of the framework.
neonpytorchheterogeneous-computingdeep-learning-inferenceheterogeneous
mwtarnowski/awesome-NeRF
Dec 2020 - Feb 2023
A curated list of awesome neural radiance fields (NeRF) papers
Contributions:22 pushes, 23 branches in 2 years 2 months
deep-learningneural-radiance-fieldsvolume-renderingneural-networknerf
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.