Ivan Novoselov
Senior Math Libraries Engineer AI And HPC
Dublin, Ireland
Summary
🤩
Rockstar🎓
Top SchoolIvan Novoselov is a Senior Math Libraries Engineer focused on AI and HPC, currently building JIT compilation solutions for cuBLAS at NVIDIA to deliver high-performance math primitives with minimal engineering overhead. With a PhD from MIPT and a background in computational materials science, he brings five years of industry experience optimizing deep learning inference stacks and low-level kernels. At Intel he contributed architecture-specific optimizations and graph-level improvements to OpenVINO, including performance-critical fixes to OneHot, ExtractImagePatches and Eltwise precision, evidencing a strong blend of systems-level engineering and numerical rigor. Comfortable in x86 assembly and JIT/codegen contexts, he enjoys squeezing more performance out of existing workloads—making them faster and cheaper—while keeping solutions maintainable and well-tested. Based in Dublin, he combines research pedigree with pragmatic production engineering across AI frameworks and math libraries.
5 years of coding experience
11 years of employment as a software developer
Doctor of Philosophy - PhD, Condensed Matter and Materials Physics, Doctor of Philosophy - PhD, Condensed Matter and Materials Physics at Moscow Institute of Physics and Technology (State University) (MIPT)
Github Skills (7)
testing10
c-language10
cprogramming-language10
performance-optimization10
ai9
deep-learning9
computer-vision9
Programming languages (1)
Github contributions (5)
openvinotoolkit/openvino
Feb 2021 - Jan 2023
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:1030 reviews, 24 commits, 132 PRs in 1 year 11 months
Contributions summary:Ivan contributed significantly to the OpenVINO toolkit by implementing and revising the OneHot operation within the Inference Engine. They added tests, optimized the ExtractImagePatches operation, and addressed Eltwise precision issues. Furthermore, the user worked on improving weights sharing synchronization, and addressing code style and serialization-related issues.
inference-enginepytorchmodel-optimizerdeep-learninggpu
IvanNovoselov/openvino
Jan 2021 - Mar 2025
OpenVINO™ Toolkit repository
Contributions:115 reviews, 14 PRs, 1069 pushes in 4 years 1 month
pytorchdeep-learninggpuopenvino-toolkitcomputer-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.