Mateusz Socha
Firmware Validation Engineer
Gdansk, Pomeranian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolMateusz Socha is a Firmware Validation Engineer with nine years of experience specializing in pre-silicon and platform-level validation for Intel, focused on power management and Optane persistent memory. He blends hands-on hardware debugging (scopes, logic analyzers, electrical measurements) with automated test development in Python and C#, and routinely authors and reviews test code and deployment tools. His open-source contributions to high-profile ML toolkits like OpenVINO, PaddlePaddle and the Nervana nGraph project show a strong backend and ML-inference interoperability skillset, including fixing ONNX import issues and adding serialization and kernel optimizations. Mateusz is adept at turning complex specs into executable test plans, submitting and triaging defects, and improving defect flow across teams. Based in Gdansk, he brings practical full-stack validation experience that bridges firmware, BIOS, OS, and ML inference stacks. Colleagues rely on him for meticulous debugging and for translating low-level platform behavior into robust, automatable test coverage.
9 years of coding experience
10 years of employment as a software developer
Universitat de Vic – Universitat Central de CatalunyaMagister (Mgr), Computer Science, Magister (Mgr), Computer Science at Uniwersytet Mikołaja Kopernika w Toruniu
Github Skills (29)
unit-testing10
serializable10
c-language10
serializer10
operation10
apidoc10
python10
tensorrt10
machine-learning10
inference10
onnx10
openvino10
serialization10
deserialization10
deeplearning-ai10
Programming languages (5)
Github contributions (5)
NervanaSystems/ngraph
Mar 2018 - May 2020
nGraph has moved to OpenVINO
Role in this project:
Back-end Developer
Back-end DeveloperContributions:156 commits, 138 PRs, 526 pushes in 2 years 2 months
Contributions summary:Mateusz primarily addressed several bugs in the codebase, focusing on fixing issues related to array flattening, padding, and call stack parameter order within the Python API of the nGraph library. The user also added unit tests for new features such as average pooling in 3D and enabled new operations like Softmax and Pad op. Furthermore, the user worked on serialization and deserialization mechanisms, enabling the API for handling various data representations.
openvinongraphtensorflowmxnetdeep-learning
openvinotoolkit/openvino
Jul 2020 - Sep 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:289 reviews, 33 commits, 41 PRs in 1 year 2 months
Contributions summary:Mateusz's contributions focus on enhancing the OpenVINO toolkit's capabilities, particularly concerning the ONNX importer and its interaction with protobuf-lite. They improved the handling of data types and tensor shapes during the import process. Additionally, the user addressed and fixed issues in the reshape builder and integrated changes to support newer ONNX versions and features. The user also worked on improvements to the model zoo runner.
aiinferenceopenvinodeep-learningcomputer-vision
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.