Igor Davidyuk
Summary
🤩
RockstarIgor Davidiuk is an AI engineer with five years of experience building scalable workflow automation and agent orchestration platforms, combining backend API design with frontend tooling to democratize AI for non-technical users. He has strong applied ML and computer vision roots from Intel and contributions to notable open-source projects like OpenFL (federated learning) and OpenVINO notebooks, where he implemented PyTorch UNet tutorials, model optimization flows and quantization pipelines. His work emphasizes privacy-preserving ML—SGX enclave training and encrypted workloads—and practical deployment concerns such as flexible input shapes and Hugging Face model integration. Based in Cyprus, Igor blends academic rigor from an MSc in accelerator physics with hands-on engineering, often translating research ideas into production-ready SDKs and interactive tutorials.
5 years of coding experience
Github Skills (12)
pytorch10
computer-vision10
machine-learning10
onnx10
federated-learning10
python10
openvino10
documentation9
inference9
deep-learning9
jupyter-notebook8
enet7
Programming languages (4)
Github contributions (5)
securefederatedai/openfl
May 2021 - Jan 2023
An Open Framework for Federated Learning.
Role in this project:
ML Engineer
ML EngineerContributions:279 reviews, 31 commits, 41 PRs in 1 year 8 months
Contributions summary:Igor primarily contributed to the development of an interactive API within the OpenFL framework, focusing on integrating and demonstrating the use of PyTorch for federated learning tasks. Their work included creating interactive API components, moving examples to tutorials, and adding extensive documentation and comments. Furthermore, the user made code changes specifically within a PyTorch UNet tutorial, implementing the model and data preparation steps, which highlights a hands-on involvement in model training within the FL context.
federated-analyticspythonsecure-computationfedavgfedopt
openvinotoolkit/openvino_notebooks
Mar 2023 - Nov 2023
📚 Jupyter notebook tutorials for OpenVINO™
Role in this project:ML Engineer
ML EngineerContributions:150 reviews, 18 PRs, 3 branches in 8 months
Contributions summary:Igor's contributions primarily focused on integrating and optimizing machine learning models within the OpenVINO™ notebooks repository. They replaced console calls with the Python API for model optimization, and integrated the Python API `convert_model()` instead of the console Model Optimizer. The user also updated the quantization API, specifically, with Python API and new NNCF API, and introduced flexible input shapes for ONNX and OpenVINO models. The user's work included implementing models from the Hugging Face hub, specifically to generate an image with text, and implementing FastSAM with OpenVINO.
deep-learningjupyter-notebooknotebookinferencecomputer-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.