Zecheng Zhang
Co-founder CTO at CAMEL-AI.org
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolZecheng Zhang is a Co-founder and CTO with eight years of experience building ML and developer tooling, currently leading TraceRoot.AI (YC S25) from Mountain View. A Stanford CS MS and former AWS AI engineer, he co-created DeepSNAP and PyTorch Frame and contributes to PyTorch Geometric, bridging graph neural network research with production systems. As a founding engineer at Kumo.AI he worked on GNNs, XAI, tabular deep learning and AutoML, and he’s a core contributor to the CAMEL multi-agent framework where he implemented OpenAI model streaming and fixed token accounting. He combines research publications and open-source maintenance with product execution, and helps run the Learning on Graphs conference. Comfortable moving between low-level C++/Python implementations and cloud-scale services, he focuses on developer-facing libraries and debuggability tools that speed root-cause analysis. His blend of academic library design and pragmatic API-level fixes reflects a rare mix of research depth and product-first engineering.
9 years of coding experience
8 years of employment as a software developer
Summer Exchange, Computer Science, Summer Exchange, Computer Science at Nanyang Technological UniversityBachelor of Science, Computer Science, Bachelor of Science, Computer Science at University of Illinois Urbana-ChampaignMaster of Science, Computer Science, Master of Science, Computer Science at Stanford University
English, Chinese
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
pytorch10
openai-api10
explainable-artificial-intelligence10
deep-learning10
graph-neural-network10
graph-convolutional-networks10
large-language-models10
python10
large-language-model10
natural-language-processing9
unit-testing9
code-coverage9
agent8
agents8
deep-learning-ai7
Programming languages (2)
Github contributions (5)
camel-ai/camel
Jun 2023 - Mar 2025
🐫 CAMEL: Finding the Scaling Law of Agents. The first and the best multi-agent framework. https://www.camel-ai.org
Role in this project:
Back-end Developer
Back-end DeveloperContributions:126 reviews, 31 PRs, 101 pushes in 1 year 9 months
Contributions summary:Zecheng contributed to improving existing examples and made minor fixes to documentation and prompts within the CAMEL framework. Their work involved refining a summarization solution extraction example, specifically modifying the `gpt_solution_extraction.py` file, showcasing a focus on improving existing functionalities. They also addressed documentation and prompt issues, indicating a role in maintaining code clarity and usability within the project. Additionally, the user enabled OpenAI model streaming and fixed a related token calculation, demonstrating knowledge of the OpenAI API and model interaction.
agentai-societiesartificial-intelligencecommunicative-aicooperative-ai
pyg-team/pytorch_geometric
Oct 2022 - Dec 2022
Graph Neural Network Library for PyTorch
Role in this project:ML Engineer
ML EngineerContributions:40 reviews, 2 commits, 32 PRs in 1 month
Contributions summary:Zecheng contributed to the PyTorch Geometric library, which is designed for graph neural networks. Their work included implementing and testing the Infection benchmark dataset for explainability algorithms and adding tests to enhance code coverage for various components such as `APPNP`, `HeteroConv`, and `PNAConv`. Furthermore, the user addressed bugs and made improvements to the `PNAConv` and `DegreeScalerAggregation` modules, and updated documentation.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
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.