Geoffrey So
Principal Machine Learning Engineer at Palo Alto Networks
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolGeoffrey So is a Principal Machine Learning Engineer with 16 years of experience applying ML, large-scale simulation, and physics-driven analytics to production systems across security, semiconductor manufacturing, and scientific computing. He builds and integrates scalable ML pipelines—ranging from NLP and deep learning for malware and email privacy at Broadcom/Symantec to real-time image analytics on Intel production lines—and has scaled training to 100+ GPUs on GCP. With a Ph.D. in Physics and deep expertise in tools like Enzo and yt, he has contributed core scientific functionality to the widely used yt project (notably halo analysis and ellipsoid calculations), bridging research-grade simulation with industrial ML deployments. Multilingual in programming and human languages, he combines strong analytical rigor with practical engineering to deliver high-impact, privacy-aware ML solutions.
16 years of coding experience
16 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Physics with specialization in Computational Science, Doctor of Philosophy (Ph.D.) Physics with specialization in Computational Science at UCSDUniversity of California, San Diego
English, Chinese, Chinese, German
Github Skills (8)
python10
astrophysics10
scientific-computing10
data-analysis10
numpy9
data-visualizations7
data-visualisation7
data-visualization7
Programming languages (4)
Github contributions (5)
yt-project/yt
Dec 2011 - Apr 2014
Main yt repository
Role in this project:
Back-end Developer
Back-end DeveloperContributions:17 commits in 2 years 4 months
Contributions summary:Geoffrey contributed significantly to the `yt-project/yt` repository, primarily focused on the `halo_objects.py` file. They made changes to the ellipsoid calculation, including fixing a missing rotation function. Further contributions involved merging fixes related to the `LoadHalo` function and updating other parts of the halo object definitions. These commits suggest involvement in core scientific computing functionality, specifically related to halo analysis and data manipulation within the yt framework.
pythondatafinite-element-analysisgeophysicsnuclear-engineering
gsiisg/xgboostGPU
Sep 2017 - Sep 2017
Contributions:24 commits, 22 pushes, 1 branch in 1 day
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.