Jiahang Xu
Research SDE 2 at 微软
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolJiahang Xu is a Research SDE 2 based in Shanghai with five years of experience building hardware-aware machine learning systems at Microsoft. He focuses on AutoML and neural architecture search, contributing notable integrations such as nn-Meter latency prediction into the popular open-source NNI toolkit to enable practical, hardware-conscious model optimization. With a master's from Fudan and an internship at Alibaba, he blends academic rigor with industry-scale engineering. He has a track record of refactoring core components (e.g., LatencyFilter) and adding support for Proxyless NAS, showing attention to both performance and reproducibility. Colleagues rely on him for turning research prototypes into robust, production-ready tooling.
5 years of coding experience
Master's degree, Master's degree at Fudan UniversityBachelor's degree, Bachelor's degree at 上海财经大学
Github Skills (6)
pytorch10
machine-learning10
automated-machine-learning10
python10
neural-architecture-search10
hyperparameter-tuning9
Programming languages (3)
Github contributions (5)
microsoft/nni
Jul 2021 - Sep 2022
An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
Role in this project:
ML Engineer
ML EngineerContributions:51 reviews, 17 commits, 21 PRs in 1 year 1 month
Contributions summary:Jiahang primarily contributed to the development and refinement of the nn-Meter integration within the NNI framework, a toolkit for automated machine learning. Their commits involved refactoring code for hardware-aware NAS, including modifications to the LatencyFilter class and the integration of nn-Meter for latency prediction. The user also refined the documentation and implemented support for the Proxyless NAS, demonstrating a focus on hardware-aware model optimization.
pythonneural-architecture-searchengineeringtensorflowbayesian-optimization
microsoft/Moonlit
Jul 2023 - Sep 2024
Contributions:28 reviews, 43 PRs, 33 pushes in 1 year 1 month
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.