Valery Chernov
Software Engineer at Nvidia
Yerevan, Armenia
Summary
🤩
RockstarValery Chernov is a software engineer and compiler specialist based in Yerevan with four years of hands-on experience building ML inference tooling. Currently contributing as a Compiler Engineer at NVIDIA on GitHub, he has deep practical experience integrating TVM into ONNX Runtime and enhancing the TVM PyTorch frontend for LSTM/GRU support. His work spans backend execution providers, VM integration, operator refactors, and practical optimizations that reduce test time and model size. Valery blends systems-level compiler knowledge with applied ML engineering to make model deployment faster and more reliable. He’s comfortable navigating large open-source ecosystems—his contributions touch high-profile projects like ONNX Runtime and Apache TVM. A less obvious strength is his focus on unifying and simplifying complex recurrent cell implementations, demonstrating both attention to performance and long-term maintainability.
4 years of coding experience
Github Skills (21)
tvm10
pytorch10
gru10
relay-modern10
machine-learning10
onnx10
deeplearning-ai10
lstm10
deep-learning10
relay10
python9
hardware-acceleration9
front-end-development9
compiler8
pytest8
Programming languages (4)
Github contributions (5)
apache/tvm
Jul 2021 - Jan 2023
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:110 reviews, 19 commits, 45 PRs in 1 year 6 months
Contributions summary:Valery contributed significantly to the PyTorch frontend of the TVM compiler stack, focusing on implementing and testing LSTM and GRU layers. Their work involved converting PyTorch models to Relay, supporting various LSTM configurations, and improving the efficiency of these operations. The user also optimized testing time by reducing model sizes and implementing alternative implementations of key PyTorch operators. Furthermore, the user unified LSTM and GRU cell implementations and extended functionality.
metalvulkancompilertensoropencl
microsoft/onnxruntime
Dec 2021 - Jul 2022
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:ML Engineer
ML EngineerContributions:73 reviews, 18 commits, 23 PRs in 7 months
Contributions summary:Valery primarily contributed to the development and enhancement of the TVM (Tensor Virtual Machine) execution provider within the ONNX Runtime project. Their work focused on integrating and improving the TVM backend, evidenced by commits related to standalone TVM executor provider support, VM integration, and optimizations. This included implementing new features, resolving build issues, and refactoring code to streamline the TVM EP's functionality.
runtimetrainingtensorflowai-frameworkaccelerator
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.