Weiyu Feng
Software Engineer at SMT (SportsMEDIA Technology)
Albany, California, United States
Summary
🤩
Rockstar🎓
Top SchoolWeiyu Feng is a software engineer and game developer with 18 years of experience focused on high-performance 3D mobile gaming and immersive XR systems. Based in Albany, CA, he blends practical Unity/C#/HLSL game development with deep research in 6D pose estimation, AR/VR training systems, and dataset creation using Azure Kinect and iPhone pipelines. He’s an active open-source contributor across major scientific and ML projects—NumPy, TensorFlow, SciPy and Dask—where his work on memory management, performance optimizations, FFT gradients and DTensor improvements demonstrates low-level systems expertise beyond typical game development. As a former GSI at UC Berkeley and VR researcher, he pairs teaching and research experience with production shipping at a sports-focused mobile game studio. Colleagues know him for making complex 3D and ML code reliable and fast, and for a rare combination of real-time graphics fluency and backend performance tuning.
18 years of coding experience
3 years of employment as a software developer
Master of Engineering - MEng, Electrical Engineering and Computer Science, Master of Engineering - MEng, Electrical Engineering and Computer Science at University of California, BerkeleyComputer Science, Computer Science at Beijing Jiaotong UniversityUniversity of Missouri
Chinese, English
Github Skills (49)
algorithm10
socket-io10
algorithms10
hdf10
debug10
dft10
c-language10
zeromq10
python10
testing10
data-structure10
numpy10
typehinting10
automatic-differentiation10
fft10
Programming languages (17)
Github contributions (5)
scipy/scipy
Jul 2015 - Aug 2019
SciPy library main repository
Role in this project:
Back-end Developer
Back-end DeveloperContributions:20 reviews, 157 commits, 18 PRs in 4 years
Contributions summary:Weiyu primarily contributed to the `scipy/scipy` repository by modifying and extending the functionality of the cKDTree class, a core component of the spatial module. Their commits include implementing periodic boundary conditions in k-NN queries, refactoring code for efficiency, and addressing issues related to memory management and potential overflow errors. The user also added weight support and extended the features of the query functions, thereby improving the overall utility of the library.
scipypythonscientific-computing
h5py/h5py
Aug 2015 - Jun 2021
HDF5 for Python -- The h5py package is a Pythonic interface to the HDF5 binary data format.
Role in this project:Backend Developer & Technical Writer
Backend Developer & Technical WriterContributions:11 commits, 4 PRs, 17 comments in 5 years 10 months
Contributions summary:Weiyu contributed to the `h5py` library by implementing bug fixes, specifically addressing vector assignment issues within compound data types. They also added a test case to verify the fix. Furthermore, the user made several revisions to the project's documentation, specifically the quick start guide, improving its clarity and flow while addressing typos and adding examples. These documentation edits focused on explaining how to read and create HDF5 files.
h5pybinary-datapythondata-formathdf5
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.