Haonan Jiang
Potsdam, Brandenburg, Germany
Summary
🤩
Rockstar🎓
Top SchoolHaonan Jiang is a Ph.D. trained remote sensing scientist with a decade of experience applying SAR and InSAR techniques to measure urban surface motion and explore bistatic imaging geometries. Trained at Wuhan University and Liaoning Technical University, he blends rigorous academic research with hands-on implementation from visiting roles at GFZ Helmholtz Centre and as an HIDA fellow. Haonan contributes to scalable machine learning infrastructure—refactoring core GBDT components in the Angel-ML parameter server—bringing practical backend and ML engineering skills to geospatial problems. Based in Potsdam, he navigates both algorithm development and production-level code, often bridging domain expertise in photogrammetry with software-driven model improvements. Colleagues value his ability to translate complex radar physics into reproducible pipelines and to integrate new loss functions and metrics that improve model accuracy.
10 years of coding experience
2 years of employment as a software developer
Ph.D, Photogrammetry and Remote Sensing, InSAR, Ph.D, Photogrammetry and Remote Sensing, InSAR at Wuhan UniversityMaster's degree, Photogrammetry and Remote Sensing, Orthorectification and Segmentation for SAR, Master's degree, Photogrammetry and Remote Sensing, Orthorectification and Segmentation for SAR at Liaoning Technical University
Github Skills (9)
gbm10
javas10
machine-learning10
machine-learning-models10
java10
breeze9
linear-algebra8
performance-optimization8
regression7
Programming languages (4)
Github contributions (5)
Angel-ML/angel
Jun 2017 - Aug 2019
A Flexible and Powerful Parameter Server for large-scale machine learning
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:257 commits, 50 PRs, 160 pushes in 2 years 2 months
Contributions summary:Haonan's commits primarily focused on refactoring and enhancing the GBDT (Gradient Boosting Decision Tree) module within the project. This involved refactoring the split-finding module, removing unnecessary logging, and improving the training parameters. The changes touched upon the core logic of the GBDT controller, indicating work on server-side functionality related to the machine learning model. The code changes also involve the integration of new loss functions and metrics for the GBDT model.
parameter-serverdeep-learningmodelmachine-learningonline-learning
bluesjjw/MyTensorFlowCode
Jul 2016 - Feb 2021
Contributions:29 pushes, 1 branch in 4 years 7 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.