Mingze Xu
Senior Staff Engineer at Ant Group
Hong Kong, China
Summary
🤩
Rockstar🎓
Top SchoolMingze Xu is a Senior Staff Engineer based in Hong Kong with eight years of experience building resilient, large-scale financial infrastructure for Ant Group’s wealth management business serving over 100 million users and managing 4,000+ billion AUM. He led the fund and asset platform R&D team and previously architected high-availability systems and QA for Alipay’s mission-critical payment campaigns like Double 11 and Spring Festival red envelopes. His background blends systems engineering, CI/CD and embedded build work with research collaborations in automated program repair, taint analysis and fuzz testing. On GitHub he has contributed to the widely used PaddlePaddle projects, optimizing TensorRT inference paths and improving CI/build support for edge platforms like BM1682 and ARM Linux. Pragmatic and research-minded, he moves smoothly between low-level performance optimizations and large-system reliability for fintech at scale. He holds a master’s in Control Science and Engineering from Zhejiang University and a top-ranked bachelor’s in Electrical and Electronics Engineering.
8 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Electrical and Electronics Engineering, 1st rank, Bachelor's degree, Electrical and Electronics Engineering, 1st rank at Harbin Engineering UniversityMaster's degree, CONTROL SCIENCE AND ENGINEERING, Master's degree, CONTROL SCIENCE AND ENGINEERING at Zhejiang University
English, Chinese
Github Skills (20)
tensorrt10
bash10
inference10
cicd10
deep-learning10
performance-optimization10
optimisation10
sys10
build-automation10
paddlepaddle10
embedded10
optimization10
c-language9
machine-learning9
cmake9
Programming languages (3)
Github contributions (5)
PaddlePaddle/Paddle
May 2021 - Jan 2023
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
Role in this project:
Back-end & ML Engineer
Back-end & ML EngineerContributions:322 reviews, 100 commits, 421 PRs in 1 year 8 months
Contributions summary:Mingze primarily contributed to the PaddlePaddle deep learning framework, specifically focusing on integrating and optimizing TensorRT for accelerated inference. Their commits involve modifying and improving the conversion process for various operators, including fully connected layers (fc), reshape, and elementwise operations to TensorRT. The user also worked on fusing quantized operations and integrating new functionalities for the transformer model generation within PaddlePaddle.
deep-learningmachine-learningpaddlepaddlescalabilityneural-network
PaddlePaddle/Paddle-Lite
Aug 2020 - Sep 2020
PaddlePaddle High Performance Deep Learning Inference Engine for Mobile and Edge (飞桨高性能深度学习端侧推理引擎）
Role in this project:DevOps Engineer & Embedded Systems Engineer
DevOps Engineer & Embedded Systems EngineerContributions:9 commits, 28 PRs, 2 comments in 23 days
Contributions summary:Mingze primarily focused on build system and CI/CD improvements within the PaddlePaddle Lite project, specifically targeting the BM1682 and ARM Linux environments. Contributions included fixing build errors related to third-party library dependencies, modifying build scripts, and adding CI implementations for the BM1682 platform, enabling testing and automated builds. Additionally, the user added example C++ demos for ARM Linux and CI.
deep-learninginference-enginemobilepaddlepaddleneural-network
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.