Zhengyao Jiang
Co-Founder & CEO at Weco AI
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolZhengyao Jiang is a PhD-trained machine learning scientist and entrepreneur who co-founded and now leads Weco AI, where he builds AI agents designed to automate the scientific innovation loop by learning from empirical feedback. With a decade of experience and deep roots in reinforcement learning from his doctoral work at UCL, he focuses on turning compute cycles into discovery through evaluation-driven search algorithms like AIDE. He combines hands-on engineering—evidenced by open-source work such as refactoring the PGPortfolio reinforcement learning repo—with product leadership to ship systems that uncover non-obvious, high-performing solutions. Based in London, he blends rigorous academic training with startup grit to pursue a bold vision: machines that not only code but generate knowledge at scale.
10 years of coding experience
1 year of employment as a software developer
Bachelor of Science - BS, Artificial Intelligence, First class, Bachelor of Science - BS, Artificial Intelligence, First class at University of LiverpoolUniversity College London
Github Skills (6)
machine-learning10
python10
tensorflow9
algorithm9
algorithms9
configuration-management8
Programming languages (6)
Github contributions (5)
ZhengyaoJiang/PGPortfolio
Nov 2017 - Jul 2021
PGPortfolio: Policy Gradient Portfolio, the source code of "A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem"(https://arxiv.org/pdf/1706.10059.pdf).
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:1 release, 30 commits, 9 PRs in 3 years 9 months
Contributions summary:Zhengyao primarily focused on refactoring and maintaining the codebase. They renamed the project's core directory from 'nntrader' to 'pgportfolio,' reflecting changes in the project's structure and dependencies. Further contributions involved fixing an activation bug and updating dependencies within the 'pgportfolio' directory. Additionally, the user modified several files, including `main.py`, `pgportfolio/learn/network.py`, and configuration files, demonstrating a hands-on approach to resolving software discrepancies.
deep-reinforcement-learningpdf
ZhengyaoJiang/GradientInduction
Mar 2018 - Apr 2019
Framework of DataLog Neural Program Synthesis
Contributions:94 commits, 10 pushes, 3 branches in 1 year
datalog
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.