Nick Korovaiko
GPU Compiler Backend Engineer at Apple
Menlo Park, California, United States
Summary
🤩
Rockstar🎓
Top SchoolNick Korovaiko is a GPU compiler backend engineer with a decade of experience building high-performance compilers and deep learning runtimes across Apple, Nuro, Meta/Facebook, Intel, and IBM. He pairs strong systems-level C/C++ expertise and compiler optimizations with practical ML framework work—contributing SymInt support to PyTorch and XLA and adding Wasm/SIMD features to ChakraCore—demonstrating fluency in both compiler IR and tensor shape semantics. His background in mathematics and an MSc in Computer Science informs a rigorous approach to graph rewrites, fusion optimizations, and code generation for accelerators. Comfortable shipping production optimizations and prototype research alike, he has driven inliner, value-propagation, and fusion work at scale and improved numeric and runtime behaviors in open-source projects. Based in Menlo Park, he blends deep technical craft with a curious, slightly irreverent open-source presence.
10 years of coding experience
15 years of employment as a software developer
MSc Computer Science, MSc Computer Science at University of VictoriaFrench Studies, French Studies at Alliance Française de TorontoBachelor Computer Science, Bachelor Computer Science at Al-Farabi Kazakh National UniversityNon credit courses French, Non credit courses French at Portland Community CollegeCertificate in French Studies, Certificate in French Studies at Seneca PolytechnicHigh School Diploma, High School Diploma at Physics and Mathematics Lyceum #166
English, French, Spanish, Russian
Github Skills (37)
pytorch10
symbolic-computation10
c-language10
operation10
python10
tensorrt10
machine-learning10
deep-learning10
tensorflow10
javascript-engine10
autograd10
xla10
simd10
compiler10
blazor-webassembly10
Programming languages (7)
Github contributions (5)
NervanaSystems/ngraph
Oct 2017 - May 2019
nGraph - open source C++ library, compiler and runtime for Deep Learning
Role in this project:
Back-end Developer
Back-end DeveloperContributions:234 commits, 215 PRs, 427 pushes in 1 year 7 months
Contributions summary:Nick primarily contributed to the nGraph library, focusing on deep learning compiler and runtime functionalities written in C++. Their commits involved implementing and testing various mathematical functions (e.g., sin, cos), debugging, and refactoring parts of the codebase. The user was also involved in integrating and optimizing operations, enhancing the library's capabilities in numerical computation and deep learning model execution.
inference-enginecppc-librarydeep-learningtvm
pytorch/pytorch
Feb 2019 - Nov 2022
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:ML Engineer
ML EngineerContributions:615 reviews, 605 commits, 599 PRs in 3 years 9 months
Contributions summary:Nick's commits primarily focused on adding support for `SymInt` in the PyTorch codebase, which involves symbolic integer representations for tensor dimensions. Their work included modifying Python bindings, implementing new methods for `SymInt` (e.g., integer division, multiplication), and integrating `SymInt` into various tensor operations like `narrow_copy`, `expand`, and `numel`. They also addressed issues related to autograd and inference, demonstrating a deep understanding of PyTorch's internals and symbolic shape representation.
pythongpu-accelerationdeep-learninggpunumpy
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.