Artur Bialas
Graphics Software Engineer at Intel Corporation
Gdańsk, Pomeranian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolArtur Bialas is a Graphics Software Engineer with nine years of experience building high-performance GPU software and deep learning inference libraries at Intel. He specializes in GPGPU, contributing both optimized primitives and graph-level optimizations for clDNN and implementing MLIR/Metal-related compiler passes in open-source projects like PlaidML. His background spans low-level kernel tuning, bug triage and validation for OpenCL, and cross-geo collaboration on driver test automation—skills that bridge research-grade algorithms and production-ready implementations. Based in Gdańsk, he pairs an M.Sc. in Electronics and Communications Engineering with practical experience exposing new operations and improving compilation and storage strategies in widely used ML compute libraries.
9 years of coding experience
7 years of employment as a software developer
Master of Science (M.Sc.), Electronics and Communications Engineering, Master of Science (M.Sc.), Electronics and Communications Engineering at Gdansk University of Technology
Polish, English, German
Github Skills (12)
opencl10
code-optimization10
c-language10
deep-learning10
deep-neural-networks10
cprogramming-language10
intel9
kernel-mode9
kernel9
spirv9
convolution9
deep-q-learning7
Programming languages (2)
Github contributions (5)
plaidml/plaidml
Aug 2020 - Jan 2021
PlaidML is a framework for making deep learning work everywhere.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:34 reviews, 23 commits, 30 PRs in 5 months
Contributions summary:Artur primarily contributed to the `plaidml/plaidml` repository by implementing and modifying passes for the Metal Language IR (MLIR) dialect. Their work focused on transforming and optimizing code, specifically related to i1 storage and subgroup operations. This included adding and modifying passes to convert i1 storage to i32 and to enable subgroup broadcast operations within the framework. The user's changes involved code modifications to C++ files and MLIR dialect files to facilitate these transformations.
pytorchtvmdeep-learningmachine-learningcompiler
intel/clDNN
May 2017 - Sep 2018
Compute Library for Deep Neural Networks (clDNN)
Role in this project:Back-end Developer
Back-end DeveloperContributions:6 commits in 1 year 3 months
Contributions summary:Artur's contributions primarily focused on fixing compilation issues and improving the overall functionality of the Compute Library for Deep Neural Networks (clDNN). They addressed layout issues, corrected engine info parameters, and exposed new operations in the eltwise API. Additionally, they modified kernels for improved performance and changed learning parameter order within the convolution depthwise separable optimized kernel.
deep-learningintelneural-networksintel-hd-graphicsneural-network
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.