Xia Chao
Tech Lead Engineer Manager at Xiaomi Technology
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolXia Chao is a tech lead and engineering manager with eight years of professional experience, currently leading teams at Xiaomi in Beijing’s Haidian District. He brings a strong foundation from roles at Baidu and Microsoft and dual degrees in Computer Science from Shanghai Jiao Tong University, including a master’s in information security. Hands-on with ML infrastructure, he has contributed to the popular MACE mobile deep learning inference framework—adding broadcasted elementwise ops, EQUAL semantics, and caffe channel-shuffle conversions with shape inference—demonstrating a blend of low-level kernel work and model portability improvements. Known for shipping pragmatic optimizations across heterogeneous mobile platforms, he combines systems thinking with practical leadership to move research-grade capabilities into product. Colleagues describe him as a developer-manager who still codes into core libraries, bridging engineering strategy and implementation.
8 years of coding experience
7 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Shanghai Jiao Tong University
Github Skills (8)
neural-network10
machine-learning10
caffe10
deep-learning10
tensorflow9
cprogramming-language9
c-language9
python8
Programming languages (2)
Github contributions (5)
XiaoMi/mace
Jun 2018 - Dec 2020
MACE is a deep learning inference framework optimized for mobile heterogeneous computing platforms.
Role in this project:
ML Engineer
ML EngineerContributions:40 commits, 4 PRs, 11 pushes in 2 years 5 months
Contributions summary:Xia contributed to the MACE deep learning inference framework by adding the MACE logo. They made changes to the `mace/kernels/eltwise.h` file, introducing the EQUAL operation, and implemented the equivalent operation in the existing TensorGeneralBroadcastEltwise, TensorBroadcastEltwise, TensorEltwise and TensorScalarEltwise methods. They also added a channel shuffle conversion for the caffe converter, along with shape inference capabilities. These changes appear to be related to new features and optimizations within the deep learning framework.
neonpytorchheterogeneous-computingdeep-learning-inferenceheterogeneous
XiaoMi/mobile-ai-bench
Jun 2018 - Sep 2019
Benchmarking Neural Network Inference on Mobile Devices
Contributions:14 commits, 1 push, 5 comments in 1 year 3 months
benchmarkingdeep-learninginferencemachine-learningbenchmark
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.