Hyunsu Cho
Senior Systems Software Engineer at NVIDIA
Las Vegas, Nevada, United States
Summary
🤩
Rockstar🎓
Top SchoolHyunsu Cho is a Senior Systems Software Engineer with a decade of experience building high-performance machine learning systems and compilers, currently at NVIDIA and long-active in open source. As lead maintainer of XGBoost since 2016, he redesigned histogram-based tree growth and optimized single-threaded performance, and helped launch Amazon SageMaker Neo during his tenure at AWS. His contributions span core ML libraries (XGBoost, cuML, TVM) and tooling for model interchange (Treelite), showing expertise across algorithm design, systems-level C++ optimization, and distributed data parsing. Hyunsu pairs research-grade algorithmic insight with pragmatic engineering—unrolling hot loops, adding dense-data specializations, and making SciPy imports conditional to increase portability. Based in Las Vegas, he holds an MS in Computer Science from the University of Washington and a dual BS in Computer Science and Mathematics, reflecting a strong theoretical foundation that informs his systems work.
10 years of coding experience
3 years of employment as a software developer
Master of Science - MS Computer Science and Engineering, Master of Science - MS Computer Science and Engineering at University of WashingtonBachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at Trinity College-Hartford
Github Skills (38)
gbm10
tvm10
markdown10
xgboost10
notebook10
c-language10
python10
technical-writing10
machine-learning10
ipython10
machine-learning-algorithms10
data-parsing10
nvidia10
regression10
compiler-compiler10
Programming languages (20)
Github contributions (5)
dmlc/treelite
Jul 2017 - Jan 2023
Universal model exchange and serialization format for decision tree forests
Role in this project:
Back-end Developer
Back-end DeveloperContributions:36 releases, 156 reviews, 547 commits in 5 years 7 months
Contributions summary:Hyunsu implemented front-end and in-memory representations, adding support for XGBoost and LightGBM formats. They focused on creating the initial architecture of the project, which included adding parser classes for xgboost.cc and implementing data structures for handling model formats. The work involved developing components for universal model exchange and serialization format, laying the groundwork for decision tree forest manipulation.
compilerdecision-treeensemblesdecision
learnml/machine-learning-specialization
Nov 2015 - Jan 2017
Role in this project:Data Scientist
Data ScientistContributions:119 commits, 27 PRs, 124 pushes in 1 year 2 months
Contributions summary:Hyunsu added blank IPython Notebook (ipynb) assignment files for the machine learning specialization course. These files provide a structure for students to complete assignments on topics such as ridge regression and LASSO. The files contain markdown and code cells, which guide students through the implementation and interpretation of these concepts within the specified course.
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.