Guohao Li is a founder and AI engineer based in the UK with eight years of experience building open-source and enterprise AI agent infrastructure. He leads CAMEL-AI.org and Eigent AI, driving development of CAMEL—a prominent multi-agent framework that explores the scaling law of agents—where he implemented core ChatAgent logic and agent orchestration. His background bridges rigorous research (PhD, postdoc at Oxford, visiting researcher at ETH Zürich) with industry experience including internships at Intel Labs and Kumo.AI working with Jure Leskovec. Practically, he has contributed key ML engineering work to PyTorch Geometric (GENConv/GenMessagePassing) and DeepGCNs, demonstrating ability to translate graph neural network research into production-ready code. He focuses on building multi-agent workforces and enterprise agents, combining system-level backend design with deep expertise in GNNs and agent-scaling — an uncommon mix of academic depth, open-source stewardship, and startup execution.
🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Role in this project:
Back-end Developer
Contributions:1 release, 447 reviews, 175 PRs in 2 years
Contributions summary:Guohao implemented and modified core functionalities within the `camel` directory, including the configuration and core logic for the `ChatAgent`. Their commits reflect a focus on defining and managing system messages, agent behavior, and tool agent structures. The user's work involved direct modifications to the underlying code structure, suggesting a deep involvement in the project's fundamental architecture and logic. They implemented critical changes and improved existing code by including new functionalities.
Pytorch Repo for DeepGCNs (ICCV'2019 Oral, TPAMI'2021), DeeperGCN (arXiv'2020) and GNN1000(ICML'2021): https://www.deepgcns.org
Role in this project:
ML Engineer
Contributions:2 reviews, 152 commits, 40 PRs in 3 years
Contributions summary:Guohao primarily focused on modifying and improving the environment setup and dependencies for the DeepGCNs project, specifically updating the required PyTorch version. They also added code and made modifications to the utility functions related to data processing and graph neural network operations, like extracting node features. Furthermore, the user integrated and modified experiments related to the OGBN-proteins dataset and simplified node feature extraction, indicating involvement in model training, dataset handling, and overall experimentation.
cheminformaticspytorchsocial-networkarxiviccv
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.