Shuai Yan
Machine Learning Engineer at Metropolis Technologies
California, United States
Summary
🤩
Rockstar🎓
Top SchoolShuai Yan is a Senior Perception Engineer with 11 years of experience building production computer vision systems, currently owning a perception stack deployed across 1,100+ sites at a $5B unicorn and powering workflows for 23,000+ employees. He specializes in 2D perception (detection, multi-object tracking, ReID, OCR) and 3D spatial intelligence (SfM, Gaussian splatting, multi-camera fusion, synthetic data), and has driven $3M+ in annual revenue uplift through model and pipeline improvements. At Metropolis he authored the core CV codebase (1,300+ commits, 90%+ ownership), raised detection mAP to state-of-the-art levels, and built synthetic-data and self-calibrating multi-camera pipelines enabling zero-shot deployments. His background spans autonomous driving and camera ISP/denoising research, and his open-source contributions include improvements to OpenMMLab segmentation tooling and localization work on the widely-used QGIS project—reflecting a practical focus on reproducible metrics and international accessibility. Based in California, he combines deep research experience with pragmatic engineering that scales from sensor noise modeling to city-scale digital twins.
11 years of coding experience
1 year of employment as a software developer
Master of Engineering - MEng, Control Engineering, Master of Engineering - MEng, Control Engineering at University of Chinese Academy of SciencesMaster of Science - MS, Computer Science, 3.97, Master of Science - MS, Computer Science, 3.97 at University of California, RiversideBachelor of Science - BS, Physics and Mathematic, Bachelor of Science - BS, Physics and Mathematic at University of Electronic Science and Technology of China
Github Skills (13)
semantic-segmentation10
computer-vision10
pytorch10
internationalization10
localization10
image-segmentation10
segmentation10
python10
deep-learning9
torchscript8
onnx8
transformer7
perl5
Programming languages (11)
Github contributions (5)
open-mmlab/mmsegmentation
Sep 2020 - Mar 2022
OpenMMLab Semantic Segmentation Toolbox and Benchmark.
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 16 commits, 8 PRs in 1 year 5 months
Contributions summary:Shuai primarily focused on improving the metrics calculation and tools related to semantic segmentation within the OpenMMLab framework. They fixed a NaN problem in the mean IoU calculation by adding an argument and also added the mFscore metric. Furthermore, the user added a tool to convert models to TorchScript format, and added model graph printing functionality to an existing configuration tool, expanding the available tools. They also made improvements to the logging of class metrics.
semantic-segmentationimage-segmentationmedical-image-segmentationbenchmarkopenmmlab
qgis/QGIS
May 2020 - May 2021
QGIS is a free, open source, cross platform (lin/win/mac) geographical information system (GIS)
Role in this project:Localization / Internationalization Specialist
Localization / Internationalization SpecialistContributions:5 commits, 3 PRs, 8 comments in 1 year
Contributions summary:Shuai's contributions primarily focused on adding and updating language translations within the QGIS project. They added a Chinese Simplified translator and subsequently updated translator names within the `tsstat.pl` script. Additionally, they corrected a spelling error in a text string within a Python file related to plots. This work indicates a focus on internationalization and ensuring the software is accessible to a wider audience.
spatiallinwmsgeotoolsgis
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.