Yang Zhang
Multimedia Engineer at ByteDance
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolYang Zhang is a multimedia engineer with 14 years of experience specializing in video codecs, DSP and ARM-optimized image processing, currently building multimedia features at ByteDance in Shanghai. He spent nearly a decade at ARM as a senior software engineer, contributing to high-performance, low-level libraries and firmware earlier in his career, and holds a master's in DSP/video codec from Zhejiang University. An active open-source contributor, he implemented CMake cross-compiling support and C++ bindings for the NE10 ARM math library and optimized NEON assembly for libyuv image scaling and YUV→RGB conversions, improving both correctness and performance. Yang combines deep algorithmic understanding with pragmatic build-system and assembly-level fixes—an unusual blend that lets him bridge research-quality signal processing and production multimedia engineering.
13 years of coding experience
11 years of employment as a software developer
master, dsp:video codec, master, dsp:video codec at Zhejiang University
Github Skills (20)
assembly10
c-language10
intrinsics10
image-processing10
build-system10
cmake10
c1110
c1710
performance-optimization10
assembler10
neon10
arm10
build-tools10
cross-compiling10
cprogramming-language10
Programming languages (2)
Github contributions (5)
projectNe10/Ne10
Jun 2012 - Jul 2014
An open optimized software library project for the ARM® Architecture
Role in this project:
Back-end Developer
Back-end DeveloperContributions:67 commits, 1 PR, 1 push in 2 years 1 month
Contributions summary:Yang primarily focused on implementing build system configurations for the "NE10" library. The contributions include adding a `CMakeLists.txt` for CMake build system and a `config.cmake` file to enable cross-compiling. The user also modified the code to support the use of NE10 in C++ programs by adding appropriate `extern "C"` declarations. Furthermore, the user was involved in fixing bugs in different functions, including `abs`, `add`, `cross`, `div`, `dot`, `mul`, `sub`, `detmat` and `invmat`, thus improving the library’s functionality and fixing the logic errors.
library-softwaresimdarmv8armv7convolution
lemenkov/libyuv
Sep 2014 - Apr 2015
Unofficial libyuv mirror. Please submit any issues or PRs upstream.
Role in this project:Back-end Developer
Back-end DeveloperContributions:17 commits in 7 months
Contributions summary:Yang primarily contributed to the `libyuv` repository by implementing and optimizing ARM64 and ARM32 NEON implementations for various image scaling and manipulation functions. They added new functions like `ScaleRowDown34_NEON`, `ScaleRowDown38_NEON`, and `ScaleAddRows_NEON`, optimizing image processing performance. They also fixed bugs and improved the accuracy of YUV to RGB conversions using NEON intrinsics. The user's work involved modifying assembly code for optimized image processing.
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.