Guohe Zheng
Collegle Professor at Ball State University
Muncie, Indiana, United States
Summary
🤩
Rockstar🎓
Top SchoolGuohe Zheng is a collegiate professor at Ball State University with 11 years of hands-on software engineering experience and a deep commitment to principled engineering and science. He blends academic rigor—holding a Ph.D. in Modern Japanese Literature—with practical ML and systems work, contributing to high-profile open-source projects like JAX, Swift for TensorFlow, and fastai. His contributions range from porting training abstractions to Swift and refining deep learning layer APIs to fixing critical bugs and improving documentation across complex compiler and numerical toolchains. Guohe also strengthens build and CI infrastructure for C++ inference engines, showing a knack for reliable deployment and DevOps. Colleagues benefit from his rare cross-disciplinary perspective: a humanities-trained researcher who writes production-level ML code and clarifying technical docs. Based in Muncie, Indiana, he quietly bridges research, teaching, and open-source engineering to make advanced tooling more usable and robust.
11 years of coding experience
Doctor of Philosophy (Ph.D.), Modern Japanese Literature, Doctor of Philosophy (Ph.D.), Modern Japanese Literature at The Ohio State University
Stackoverflow
1,593reputation
44kreached
17answers
7questions
command-line-interface
top-5%
Github Skills (37)
github-ci10
python10
machine-learning10
cicd10
numpy10
deep-learning10
tensorflow10
mnist10
build-automation10
swift10
differentiable-programming10
jax10
githubaction-workflow10
haskell10
documentation10
Programming languages (19)
Github contributions (5)
tensorflow/swift
Mar 2019 - Feb 2021
Swift for TensorFlow
Role in this project:
Technical Writer & Code Maintainer
Technical Writer & Code MaintainerContributions:17 reviews, 69 commits, 199 PRs in 1 year 11 months
Contributions summary:Guohe primarily focused on improving the documentation and fixing issues within the Swift for TensorFlow project. Their contributions include fixing broken links and quotation marks, escaping quotation marks in tutorials, and moving the license information to the top of the tutorial files. They also updated the tutorials to reflect changes in API usage, specifically regarding the removal of deprecated method-style differential operators. These updates suggest a role focused on maintaining and refining the user-facing documentation and examples within the repository.
differentiable-programmingswift-for-tensorflowmachine-learningswifttensorflow
tensorflow/swift-models
Apr 2018 - Jan 2021
Models and examples built with Swift for TensorFlow
Role in this project:ML Engineer
ML EngineerContributions:16 reviews, 64 commits, 73 PRs in 2 years 9 months
Contributions summary:Guohe primarily contributed to the development of a Swift-based machine-learning model for the MNIST dataset within the Swift for TensorFlow ecosystem. Their work focused on data loading, model implementation, and training loop adjustments. They debugged and improved data handling, optimized the training process by utilizing Foundation APIs, and implemented mini-batch gradient descent. The user also refactored the model using updated APIs, corrected loss calculations, and added minor improvements.
swifttensorflowswift-for-tensorflow
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.