Xiang An
Software Engineer at Facebook
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top School
Top expert inArtificial Intelligence and Computer Vision Technologies
Xiang An is a software engineer with eight years of experience building large-scale search and machine learning systems, currently based in Mountain View and working at Facebook. He has deep search-infrastructure roots from multiple roles at LinkedIn and SRCH2, and now applies that systems expertise to back-end ML engineering. His open-source contributions include evaluation and training improvements to the well-known InsightFace face-analysis project and practical training refinements for remote-sensing segmentation models, reflecting a focus on model evaluation, optimization, and reproducible training. Xiang combines production-grade engineering with hands-on ML work—optimizing evaluation scripts, integrating gradient checkpointing, and tuning training configs. He holds a B.S. from Shanghai Jiao Tong University and a Master's in Computer Science from UC Irvine, blending strong academic foundations with production experience. Colleagues would describe him as a pragmatic problem-solver who bridges search-scale systems thinking with applied deep learning.
8 years of coding experience
6 years of employment as a software developer
Master, Computer Science, Master, Computer Science at University of California, IrvineThe High School Affiliated to Renmin University Of ChinaB.S., Computer Science, B.S., Computer Science at Shanghai Jiao Tong University
Chinese
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
mxnet10
pytorch10
machine-learning10
trainings10
tensorflow10
python10
evaluation10
modeling10
face-recognition10
faster-rcnn9
mask-rcnn9
computer-vision9
Programming languages (1)
Github contributions (5)
anxiangsir/urban_seg
Apr 2018 - Jan 2022
Remotes Sensing Semantic Segmentation
Role in this project:
ML Engineer
ML EngineerContributions:38 commits, 4 PRs, 134 pushes in 3 years 9 months
Contributions summary:Xiang primarily focuses on modifying the `train.py` file within the remote sensing semantic segmentation project. Their contributions revolve around adapting the training process, specifically by controlling the use of pre-trained ResNet models and modifying hyperparameter settings for model training. This also includes updating dependencies and parameters. The user is making improvements to training and model configuration within the provided deep learning framework.
deeplabdeeplabv3python3remote-sensingtensorflow
deepinsight/insightface
Sep 2020 - Dec 2022
State-of-the-art 2D and 3D Face Analysis Project
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:234 commits, 23 PRs, 323 pushes in 2 years 2 months
Contributions summary:Xiang primarily updated and modified code related to the evaluation of face recognition models within the insightface repository. Their contributions centered on enhancing the evaluation scripts, specifically within the context of the IJB (IJB-C and IJB-B) datasets. They also made changes related to the partial FC implementation, including modifications to configurations, training settings, and incorporated gradient checkpointing. These changes indicate a focus on model testing, optimization, and potentially model training.
face-recognitionretinafacestate-of-the-artface-analysisarcface
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.