Xuhang Cao
Python Developer at BeaconFire Inc.
Atlanta, Georgia, United States
Summary
🤩
Rockstar🎓
Top SchoolXuhang Cao is a Python developer with a decade of hands-on experience building backend systems, research prototypes, and deployment tooling from Atlanta. He has contributed to high-profile open-source projects like PyTorch and the EVA database, improving export serialization and adding Docker support that made an AI-powered database easier to deploy on CPU/GPU environments. Xuhang’s background blends academic research—optimizing image classification and power-flow solvers at CMU and Georgia Tech—with production-facing roles, including internships at IBM and a full-stack position at Flomad Labs. He is comfortable across the stack: writing clean backend code, creating containerized deployments, and integrating ML components into services and tests. Notably, he turned large simulation datasets and nuanced termination criteria into actionable models, demonstrating skill at turning research into reliable engineering. He holds a CS BA from Oberlin and is pursuing graduate studies at Georgia Tech, pairing rigorous research experience with practical delivery.
10 years of coding experience
2 years of employment as a software developer
Master's degree, Master's degree at Georgia Institute of TechnologyBachelor of Arts - BA, Computer Science, Bachelor of Arts - BA, Computer Science at Oberlin College
Github Skills (23)
pytorch10
docker10
python10
databases10
machine-learning10
dockers10
deep-learning10
database10
ai9
neural-network9
tensor9
object-detection9
test-automation8
autograd8
mysql8
Programming languages (10)
Github contributions (5)
georgia-tech-db/evadb
Oct 2020 - Jan 2023
Database system for AI-powered apps
Role in this project:
Backend & DevOps Engineer
Backend & DevOps EngineerContributions:1 release, 282 reviews, 71 commits in 2 years 4 months
Contributions summary:Xuhang's commits primarily focused on integrating Docker support for the EVA database system. This included creating Dockerfiles for CPU and GPU environments, establishing a Docker Compose setup for testing, and modifying configurations to integrate with MySQL. The user also added and tested an SSD object detector, implementing tests and resolving related indexing errors. Overall, the contributions were crucial in making EVA AI-powered apps more accessible and easier to deploy.
image-classificationvideogpt-4evaexploratory-video-analytics
pytorch/pytorch
May 2024 - Aug 2024
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:Back-end Developer
Back-end DeveloperContributions:33 reviews, 45 PRs, 207 pushes in 2 months
Contributions summary:Xuhang's contributions focused on refactoring and cleaning up serialization code within the PyTorch framework, specifically targeting the `_export` module. They removed redundant code paths related to handling different types in `nn_module` metadata, improving code clarity and maintainability. Further work included refactoring variables and function names, improving readability, and separating non/strict export functions within the `_export` module. These changes suggest a focus on improving the internal workings and efficiency of the PyTorch export functionality.
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.