Zhanghan Ke
Hong Kong, China
Summary
🤩
Rockstar🎓
Top SchoolZhanghan Ke is a Ph.D. candidate in Computer Science at CityU Hong Kong specializing in data-efficient AI and image/video processing, with a decade of industry and research experience spanning Microsoft, SenseTime, and Adobe. He has built practical algorithms for matting and color tuning that have been adopted by numerous companies worldwide and is focused on creating AI-driven editing tools that make image and video manipulation more intuitive. His open-source contributions include enhancements to Microsoft’s CNTK deep-learning toolkit, demonstrating low-level ML engineering skills in C++ and API design. Combining rigorous academic research with production-oriented internships, he’s open to collaborations that translate advanced self-/semi-supervised learning into real-world multimedia applications.
10 years of coding experience
Bachelor of Engineering - BE, Software Engineering, Bachelor of Engineering - BE, Software Engineering at Northeastern University (CN)Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at City University of Hong Kong
Github Skills (8)
machine-learning10
c-language10
cntk10
deep-learning10
cprogramming-language10
python9
api8
apidoc8
Programming languages (3)
Github contributions (5)
microsoft/CNTK
Mar 2017 - Aug 2017
Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
Role in this project:
ML Engineer
ML EngineerContributions:54 commits, 12 PRs, 33 pushes in 5 months
Contributions summary:Zhanghan primarily contributed to the Microsoft Cognitive Toolkit (CNTK), a deep-learning toolkit, by modifying and adding support for average pooling layers. These changes included modifications to C++ headers and source files within the core computation network library to add the "includePad" parameter to the average pooling layer and integrating this new parameter into the BrainScript definition. Furthermore, the user fixed bugs related to the average pooling with includePad in the BS script and renamed the poolPadMode parameter to includePad in the Python API. These changes indicate the user's involvement in the refinement and feature addition to core functionalities.
pytorchpythondeep-learningc-plus-plusmachine-learning
ZHKKKe/DualStudent
Sep 2019 - Aug 2020
Contributions:15 commits, 11 pushes, 1 branch in 11 months
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.