Boris Knyazev
Adjunct Professor at Université de Montréal
Montreal, Quebec, Canada
Summary
👤
Senior🎓
Top SchoolBoris Knyazev is a research scientist and adjunct professor based in Montreal with a decade of experience at the intersection of machine learning and computer vision. He currently leads research at Samsung AI Lab Montreal and teaches at Université de Montréal, bringing deep academic rigor from a PhD and a track record of internships at top labs including Facebook AI and Mila. His contributions to the PyTorch Geometric project and work on graph neural networks reflect hands-on open-source impact beyond traditional vision tasks. Earlier industry roles delivered top-tier results in face and emotion recognition competitions, demonstrating an ability to translate research into high-performing systems. Known for blending classical image-processing expertise with modern deep learning, he often bridges theory and practical deployment in large-scale research settings.
10 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD, ENGINEERING, Doctor of Philosophy - PhD, ENGINEERING at University of GuelphEngineer’s Degree, Information Technology, 4.8 / 5.0, Engineer’s Degree, Information Technology, 4.8 / 5.0 at Bauman Moscow State Technical University
English, French
Github Skills (7)
pytorch10
machine-learning10
deep-learning10
pytorch-geometric10
graph-neural-network10
graph-convolutional-networks10
python10
Programming languages (3)
Github contributions (5)
pyg-team/pytorch_geometric
Dec 2018 - Jan 2020
Graph Neural Network Library for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:9 commits, 3 PRs, 6 comments in 1 year
Contributions summary:Boris contributed to the PyTorch Geometric library, primarily focusing on features related to graph neural networks. Their work includes implementing a flag for using continuous node attributes in the `TUDataset`, adding examples for `COLORS` and `TRIANGLES` datasets, and adjusting pooling layers such as `TopKPooling` and `SAGPooling`. The commits also include pycodestyle formatting and addition of a transformer for node attention.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
facebookresearch/ppuda
Oct 2021 - Jul 2022
Code for Parameter Prediction for Unseen Deep Architectures (NeurIPS 2021)
Contributions:20 commits, 4 PRs, 8 comments in 9 months
pytorcharchitecturespredictionneurips-2021unseen
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.