Weikang Song
Staff Software Engineer at Circle
Kirkland, Washington, United States
Summary
👤
Senior🎓
Top SchoolWeikang Song is a Staff Software Engineer based in Kirkland with 11 years building scalable systems across Google, Microsoft, and fintech at Circle, combining deep expertise in Python, C++, OCR and pattern recognition with a Master’s in Computer Science from Peking University. He has driven production-grade ML and distributed-computation work — notably contributing serialization, type annotations, and testing enhancements to the TensorFlow Federated project to make decentralized computations more robust and maintainable. Comfortable moving between research-level problems and pragmatic engineering, he has a track record of shipping APIs and migration work that reduce technical debt and enable new use cases. Colleagues rely on him for clean abstractions and durable systems design informed by both industry-scale product demands and academic rigor.
11 years of coding experience
9 years of employment as a software developer
Bachelor of Engineering, Electronic Engineering, Bachelor of Engineering, Electronic Engineering at Beijing University of Posts and TelecommunicationsMaster of Engineering (M.Eng.), Computer Science, Master of Engineering (M.Eng.), Computer Science at Peking University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
deserialization10
machine-learning10
serializable10
tensorflow10
serializer10
federated-learning10
python10
serialization10
type-annotations9
apidoc9
api9
testing8
Programming languages (4)
Github contributions (5)
google-parfait/tensorflow-federated
Apr 2020 - Mar 2021
An open-source framework for machine learning and other computations on decentralized data.
Role in this project:
ML Engineer
ML EngineerContributions:18 commits in 10 months
Contributions summary:Weikang primarily focused on enhancing the serialization and deserialization capabilities for computations within the TensorFlow Federated framework. They implemented serialization/deserialization functionalities for `tff.Computation` and added type annotations to improve code maintainability. The user also contributed to improvements within the type serialization library and added testing to validate functionality. They further contributed to the development of APIs for creating `MeasuredProcess` objects based on models and encoders, migrating use cases.
pytorchdeep-learningmachine-learningsecure-computationfederated-learning
swkpku/Detectron
Mar 2018 - Mar 2018
Contributions:2 PRs, 25 pushes, 4 branches in 7 days
pytorchmaskdeep-learningretinanetr-cnn
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.