Po-yu Kao
Silicon Engineer at Google
New Taipei, Taiwan
Summary
👤
Senior🎓
Top SchoolPo-yu Kao is a Silicon Engineer at Google with a Ph.D. in Electrical Engineering from UC Santa Barbara and nine years of experience applying machine learning and generative models to real-world problems. He has driven drug discovery and DTI advances at Insilico Medicine—leading EnsembleDLM to state-of-the-art results—and contributed to widely used open-source bioinformatics tooling like DeepPurpose. His background blends deep technical rigor in medical image analysis and chemistry-focused generative models with hands-on engineering: debugging core data pipelines, adding novel encoders, and integrating monitoring like TensorBoard. Comfortable leading teams and collaborating with industry partners, he pairs strong academic credentials and publishing experience with practical delivery across startups and large tech. An unspoken strength is his knack for translating complex research into production-ready ML systems that directly impact drug discovery workflows.
9 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy - PhD, Electrical Engineering, 3.94/4.0, Doctor of Philosophy - PhD, Electrical Engineering, 3.94/4.0 at UC Santa BarbaraBachelor of Science - BS, Electrical Engineering, 3.51/4.0, Bachelor of Science - BS, Electrical Engineering, 3.51/4.0 at National Taipei University of Technology
Chinese, English, Mandarin
Stackoverflow
2,647reputation
475kreached
4answers
6questions
Github Skills (23)
pytorch10
python10
machine-learning10
deep-learning10
bioinformatics10
data-processing9
tensorboard9
drug-discovery8
numpy8
api-design7
apim7
pandas7
api7
nlp6
wxpython6
Programming languages (4)
Github contributions (5)
kexinhuang12345/DeepPurpose
Nov 2020 - Feb 2021
A Deep Learning Toolkit for DTI, Drug Property, PPI, DDI, Protein Function Prediction (Bioinformatics)
Role in this project:
ML Engineer
ML EngineerContributions:12 commits, 18 PRs, 20 comments in 2 months
Contributions summary:Po-yu primarily contributed to the DeepPurpose toolkit, focusing on its functionality for drug-target interaction (DTI) prediction and related bioinformatics tasks. Their work involved debugging existing code, particularly in the `utils.py` file related to data processing and unit conversion, and updating the BindingDB dataset to the latest version. They enhanced the toolkit by adding an ErG decoder for drug encoding and integrated TensorBoard for tracking the training process of DTI models, demonstrating a focus on model development, optimization, and evaluation.
drug-target-interactionsprotein-functionbioinformaticsdrugchainer
pykao/BraTS2018-tumor-segmentation
Sep 2018 - Jul 2020
Contributions:79 commits, 27 pushes, 1 branch in 1 year 10 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.