Yingkai Sha
Project Scientist at NSF NCAR - The National Center for Atmospheric Research
Boulder, Colorado, United States
Summary
👤
Senior🎓
Top SchoolYingkai Sha is a Project Scientist at NCAR with 11 years of experience applying numerical weather prediction and AI to atmospheric science problems. He holds a Ph.D. in Atmospheric Science from UBC and has blended academic research with operational projects—developing bias-correction, calibration, and downscaling models for precipitation while building near-real-time quality-control systems. Yingkai is an active ML engineer in open source, contributing substantial improvements to a popular Keras U-Net collection (adding TransUNET, deep supervision, and novel loss integrations) that bridges research models and practical deployments. Based in Boulder, he combines strong statistical and deep-learning expertise with hands-on verification of extreme-event forecasts, making him effective at turning cutting-edge methods into stakeholder-ready products.
11 years of coding experience
8 years of employment as a software developer
Nanjing University of Information Science and TechnologyDoctor of Philosophy - PhD, Atmospheric Science, Doctor of Philosophy - PhD, Atmospheric Science at The University of British Columbia
Github Skills (9)
net10
computer-vision10
machine-learning10
image-segmentation10
deep-learning10
segmentation10
tensorflow10
modeling10
implement10
Programming languages (4)
Github contributions (5)
yingkaisha/keras-unet-collection
Dec 2020 - Jul 2022
The Tensorflow, Keras implementation of U-net, V-net, U-net++, UNET 3+, Attention U-net, R2U-net, ResUnet-a, U^2-Net, TransUNET, and Swin-UNET with optional ImageNet-trained backbones.
Role in this project:
ML Engineer
ML EngineerContributions:24 releases, 219 commits, 2 PRs in 1 year 6 months
Contributions summary:Yingkai's primary contributions focused on implementing and improving various U-Net-based models within the repository. They added new model implementations like transUNET and updated existing models such as the base UNet and U^2-Net. The commits indicate the addition of new features related to deep supervision and the integration of new loss functions, showing a focus on model architecture enhancements and experimentation.
pypikeras-tensorflowswinu-nettensorflow
yingkaisha/JAMC_20_0057
Dec 2020 - May 2021
Contributions:15 commits, 12 pushes, 1 branch in 5 months
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.