Mikołaj Życzyński
AI Software Architect at Intel Corporation
Grudziądz, Kuyavian-Pomeranian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolMikołaj Życzyński is an AI software architect with six years of experience designing and optimizing production-grade ML inference systems at Intel. He progressed from software development engineer to technical lead for an enterprise RAG project and now architects AI solutions, blending hands-on optimization with project leadership. His open-source contributions to the widely used OpenVINO toolkit demonstrate deep expertise in performance tuning—debugging grouped convolutions, fused activations, quantization fixes, and kernel improvements for real-world inference workloads. Trained as a robotics and control engineer with a background in power engineering from Gdańsk University of Technology, he brings strong systems thinking and low-level optimization skills to model deployment. Based in Grudziądz, Poland, he pairs pragmatism with a knack for squeezing latency and accuracy gains out of existing runtimes. Notably, he moves comfortably between kernel-level fixes and higher-level RAG architecture, making him effective at closing the gap between research models and robust production inference.
5 years of coding experience
5 years of employment as a software developer
Master of Engineering - MEng, Robotics and automatic control, Master of Engineering - MEng, Robotics and automatic control at Gdańsk University of Technology
Github Skills (9)
computer-vision10
deeplearning-ai10
c-language10
deep-learning10
cprogramming-language10
openvino10
optimization10
ai9
inference9
Programming languages (2)
Github contributions (5)
openvinotoolkit/openvino
May 2020 - Oct 2020
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference
Role in this project:
ML Engineer
ML EngineerContributions:2 reviews, 9 commits, 9 PRs in 5 months
Contributions summary:Mikołaj primarily focused on fixing bugs and improving the performance of the OpenVINO toolkit. Their contributions involved debugging and optimizing grouped convolution operations within the clDNN backend. They also addressed issues related to activation functions and quantization, implementing fused operations and adding corresponding tests. Further work included improvements to the reduce and pooling kernels.
inference-enginepytorchmodel-optimizerdeep-learninggpu
mzyczyns/openvino
May 2020 - Oct 2020
OpenVINO™ Toolkit - Deep Learning Deployment Toolkit repository
Contributions:75 pushes, 9 branches in 5 months
pytorchdeep-learningdeploymentopenvino-toolkitobject-detection
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.