Bartosz Myrcha
AI Software Solutions Engineer at Intel Corporation
Gdansk, Pomeranian Voivodeship, Poland
Summary
🤩
RockstarBartosz Myrcha is an AI Software Solutions Engineer based in Gdansk with five years of hands-on experience building and productionizing ML tooling. Currently at Intel, he focuses on back-end and DevOps aspects of model compression and quantization workflows, contributing fixes and integrations across MXNet, TensorFlow, PyTorch, and ONNX Runtime. His open-source work on the high-profile intel/neural-compressor project shows attention to code quality, cross-framework interoperability, and practical deployment concerns like logging and examples. Comfortable bridging research-oriented model techniques and engineering-grade pipelines, he brings pragmatic problem-solving to performance-sensitive AI systems. Colleagues can expect a detail-oriented engineer who combines low-level adaptor tweaks with an eye toward reproducible, scalable model delivery.
6 years of coding experience
English, Polish
Github Skills (5)
quantization10
python10
tensorflow9
mxnet9
pylint8
Programming languages (1)
Github contributions (5)
intel/neural-compressor
Jul 2020 - Dec 2022
SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime
Role in this project:
Back-end Developer & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:23 reviews, 54 commits, 20 PRs in 2 years 5 months
Contributions summary:Bartosz's commits primarily involved code adjustments and adherence to pylint rules within the MXNet adaptor, indicating a focus on the back-end components related to model quantization. These changes included modifications to the MXNet adaptor, metric definitions, and strategy implementations. Additionally, the user addressed a TensorFlow typo in logging and updated the helloworld TF example, showcasing interaction with multiple deep learning frameworks and potentially some build/release contributions.
llmmodel-compressiononnxruntimepytorchquantization
bmyrcha/pyg-lib
May 2023 - Jun 2023
Low-Level Graph Neural Network Operators for PyG
Contributions:96 pushes, 3 branches in 14 days
graph-neural-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.