Bartosz Sochacki
Software Architect at Intel Corporation
Poland
Summary
🤩
Rockstar🎓
Top SchoolBartosz Sochacki is a Software Architect with over a decade at Intel, evolving from intern to senior roles in graphics and architecture while applying an MSc in Technical Physics and Applied Informatics. He specializes in back-end and ML engineering for inference tooling, contributing notable fixes and feature work to OpenVINO’s GNA plugin (including power-layer support and int8/int16 fake quantization) and to Khronos OpenCL headers to improve cross-vendor compatibility. Comfortable in low-level systems and performance-sensitive code, he blends hardware-aware engineering with practical ML deployment experience. Based in Poland, he brings a steady track record of shipping robust, standards-conscious contributions to high-profile open-source projects.
10 years of coding experience
MSc, Technical Physics and Applied Mathematics - Applied Informatics (Computer science), MSc, Technical Physics and Applied Mathematics - Applied Informatics (Computer science) at Politechnika Gdańska / Technical University of Gdansk
Stackoverflow
66reputation
6kreached
2answers
0questions
Github Skills (14)
c1710
header-files10
opencl10
c-language10
inference10
cprogramming-language10
c1110
ai10
openvino10
optimization10
deep-learning9
computer-vision8
pyopencl6
atomic6
Programming languages (6)
Github contributions (5)
KhronosGroup/OpenCL-Headers
Sep 2015 - Sep 2016
Khronos OpenCL-Headers
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 commits, 5 PRs, 1 comment in 11 months
Contributions summary:Bartosz primarily contributed to the OpenCL-Headers repository by addressing various bugs and inconsistencies within the header files. Their work involved correcting typedefs, adding missing definitions related to OpenCL versions and data types, and fixing incorrect macro definitions. The changes focused on ensuring the accuracy and completeness of the header files for OpenCL, a crucial step for developers using the OpenCL standard. They also added vendor-specific headers for Intel extensions.
headerskhronosopencl
openvinotoolkit/openvino
Jul 2020 - Mar 2021
OpenVINO™ is an open source toolkit for optimizing and deploying AI inference
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:168 reviews, 6 commits, 25 PRs in 7 months
Contributions summary:Bartosz's primary contributions focus on enhancing the GNA plugin for the OpenVINO toolkit, specifically addressing power layer support and implementing fake quantization layers. They added support for power layers with non-1 exponents, and resolved issues related to concat layers with multiple inputs. The user implemented improvements for int8 and int16 fake quantization. Moreover, they refined the code related to scale factor calculations and fixed various code style issues.
inference-enginepytorchmodel-optimizerdeep-learninggpu
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.