Hoe Yang
System Software Development Engineer at Intel Corporation
Penang, Malaysia
Summary
👤
Senior🎓
Top SchoolHoe Yang is a System Software Development Engineer with 15 years of experience and over 7 years focused on software and media stack development at Intel, specializing in C/C++ and driver-to-user-space solutions for Intel GPUs and VPUs. He has owned end-to-end media software components—from kernel drivers and libdrm/libva to gstreamer plugins and oneVPL automation—delivering hardware-accelerated features, performance optimizations, and CI/CD-driven validation. An active contributor to high-profile open-source projects, he has modernized Intel AI reference models to TensorFlow 2.0 and improved TensorFlow's C API kernel handling, demonstrating cross-domain expertise in ML integration and low-level systems. Known for strong analytical troubleshooting and mentorship of junior engineers, he balances deep technical ownership with attention to compliance and release-quality metrics. Based in Penang, Malaysia, he combines academic rigor (First Class Honours in Computer Engineering) with a pragmatic, learn-from-failure approach to complex platform engineering.
15 years of coding experience
1 year of employment as a software developer
Bachelor's degree, Computer Engineering, First Class Honours CGPA 3.91, Bachelor's degree, Computer Engineering, First Class Honours CGPA 3.91 at Universiti Malaysia Perlis
English, Chinese, Malay
Github Skills (14)
machine-learning10
c-language10
inference10
api10
tensorflow10
deep-learning10
cprogramming-language10
python10
apidoc10
computer-vision8
ai8
deep-neural-networks7
cpu7
neural-network7
Programming languages (5)
Github contributions (5)
tensorflow/tensorflow
Aug 2019 - Jul 2021
An Open Source Machine Learning Framework for Everyone
Role in this project:
Back-end Developer
Back-end DeveloperContributions:21 reviews, 54 commits, 27 PRs in 2 years
Contributions summary:Hoe's contributions focused on improving the TensorFlow C API, specifically regarding kernel building and execution. They introduced the ability to reuse the `TF_KernelBuilder` for both synchronous and asynchronous kernels. The user also fixed a bug related to the CApi Bypass path and addressed issues with the NextPluggableDevice's allocator and synchronization. Furthermore, the user implemented recursive handling for the `AddN` variant.
pythondata-sciencedeep-learningmlmachine-learning
intel/ai-reference-models
Jul 2019 - Dec 2019
Intel® AI Reference Models: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors and Intel® Data Center GPUs
Role in this project:ML Engineer
ML EngineerContributions:7 commits in 4 months
Contributions summary:Hoe's contributions primarily involve upgrading and adapting existing code within the Intel AI Reference Models repository to TensorFlow 2.0. This includes changes across multiple files, indicating a broad impact on model compatibility. The modifications demonstrate expertise in adjusting the image recognition and object detection models for the new TensorFlow version. Additionally, the user addresses int8 accuracy issues.
optimizationsprocessorstensorflowzoomodel-zoo
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.