Wennan Zhu
Machine Learning Scientist at Apple
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolWennan Zhu is a machine learning scientist with a decade of experience bridging theoretical research and production systems in privacy-preserving analytics and auction algorithms. Currently at Apple after research and product roles at Google, he has advanced federated analytics, differential privacy, and auction optimization for major ad products while contributing open-source fixes to TensorFlow Federated’s TrieHH heavy-hitter implementation. His background spans a PhD in computer science and earlier work in molecular biology, reflecting interdisciplinary rigor and an ability to translate deep theory into deployable systems. Known for tackling prefix-handling and vote-accumulation bugs that improved heavy-hitter estimation performance, he combines algorithmic game theory insights with hands-on backend engineering to deliver scalable, privacy-aware solutions.
10 years of coding experience
14 years of employment as a software developer
Master of Science (MS), Molecular Biology, Master of Science (MS), Molecular Biology at Peking Union Medical CollegeDoctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at Rensselaer Polytechnic InstituteBachelor of Science (BS), Computer Science and Technology, Bachelor of Science (BS), Computer Science and Technology at Tsinghua University
Github Skills (5)
tensorflow10
python10
federated-learning10
trie10
machine-learning8
Programming languages (2)
Github contributions (5)
google-parfait/tensorflow-federated
Jul 2020 - Aug 2022
An open-source framework for machine learning and other computations on decentralized data.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 release, 37 commits, 7 PRs in 2 years
Contributions summary:Wennan primarily focused on fixing bugs and improving the functionality of the `TrieHH` (Trie Heavy Hitter) component within the TensorFlow Federated framework. They addressed issues related to prefix handling and the accumulation of client votes, ensuring correct behavior when handling strings with prefixes. Additionally, the user updated the heavy hitters estimation, and modified the code to improve overall performance. These modifications involved changes to the core logic of the TrieHH implementation and associated tests.
pytorchdeep-learningmachine-learningsecure-computationfederated-learning
triehh/triehh
Feb 2020 - Nov 2020
Code for Federated Heavy Hitters with Differential Privacy NeurIPS2019 submission.
Contributions:3 commits, 1 PR, 3 pushes in 8 months
privacydifferential-privacydifferentialheavy-hittersfederated
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.