Alexander Dokuchaev
Deep Learning R&D Engineer at Intel Corporation
Nizhniy Novgorod, Nizhniy Novgorod, Russia
Summary
🤩
RockstarAlexander Dokuchaev is a Deep Learning R&D Engineer with 11 years of experience, currently advancing model optimization and inference at Intel in Nizhniy Novgorod. He contributes to high-impact OpenVINO projects—implementing filter pruning for ConvTranspose layers in the Neural Network Compression Framework and extending super-resolution demos and LPR model compatibility in the Open Model Zoo. His work blends practical engineering and research-driven improvements, focusing on reducing FLOPs and enhancing deployment-ready performance. Known for fixing subtle pruning bugs and adapting demos to diverse image inputs, he brings a pragmatic eye for production constraints and cross-model integration.
11 years of coding experience
Github Skills (15)
compression10
computer-vision10
pruning10
pytorch10
cnn-model10
deep-learning10
modello10
openvino10
model-optimization10
cprogramming-language9
c-language9
inference9
onnx7
quantization7
tensorflow5
Programming languages (3)
Github contributions (5)
openvinotoolkit/nncf
Nov 2020 - Jan 2023
Neural Network Compression Framework for enhanced OpenVINO™ inference
Role in this project:
ML Engineer
ML EngineerContributions:1006 reviews, 37 commits, 520 PRs in 2 years 2 months
Contributions summary:Alexander primarily contributed to the Neural Network Compression Framework, focusing on filter pruning techniques. Their commits involved implementing and refining filter pruning functionality, specifically for ConvTranspose layers. They also made contributions related to calculating and displaying FLOPs pruning levels, indicating a focus on model optimization and performance analysis within the context of model compression. This user worked on fixing bugs and enhancements for the pruning algorithms.
bertsemantic-segmentationmixed-precision-trainingtensorflowclassification
openvinotoolkit/open_model_zoo
Aug 2019 - Nov 2019
Pre-trained Deep Learning models and demos (high quality and extremely fast)
Role in this project:ML Engineer
ML EngineerContributions:16 commits, 6 PRs, 18 comments in 2 months
Contributions summary:Alexander primarily contributed to the "super_resolution_demo" within the Open Model Zoo repository. They implemented models, modified image loading and processing functionalities, and updated the codebase to accommodate different image input formats. Their contributions included adding a text-image-super-resolution model and incorporating compatibility with LPR (License Plate Recognition) models from OTE (OpenVINO Training Extensions). This indicates a focus on integrating and extending model capabilities within the existing demonstration framework.
modelonnx-modelsdeep-learning-modelsmodel-zoopytorch-models
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.