Zhiwei Chen
Sr. Machine Learning Engineer at Roku Inc.
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolZhiwei Chen is a Senior Machine Learning Engineer in San Jose with 12 years of software and ML experience, currently focused on large-scale news recommendation and personalization for The Roku Channel. He designs hybrid recommender systems that fuse interaction signals, user and item features, and contextual data—drawing on ideas from StarSpace and Wide&Deep—to deliver measurable CTR gains across 3+ million users while handling cold-starts effectively. Comfortable taking models from research to production, he writes PyTorch models, Spark pipelines, and SOA deployment services, and has built offline evaluation and A/B testing frameworks to iterate quickly. His background spans distributed systems and high‑throughput architectures at Yihaodian and Huawei, giving him a rare combination of ML modeling depth and production reliability. He’s also contributed backend enhancements to the widely used WikiExtractor tool, improving category filtering, UTF-8 handling, and logging for large-scale text extraction.
12 years of coding experience
10 years of employment as a software developer
Master’s Degree, Computer Science, Master’s Degree, Computer Science at Florida State UniversityBachelor’s Degree, Microeletronics, Bachelor’s Degree, Microeletronics at Sun Yat-Sen University
English, Chinese
Github Skills (9)
regular-expression10
python10
text-processing10
fileio9
file-access9
data-filtering9
file-handling9
file-processing9
logging8
Programming languages (4)
Github contributions (5)
WikiExtractor/wikiextractor
Apr 2017 - Apr 2017
A tool for extracting plain text from Wikipedia dumps
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 commits, 1 PR in 2 days
Contributions summary:Zhiwei primarily focused on enhancing the functionality of the WikiExtractor tool. Their contributions include adding features for filtering Wikipedia content based on category inclusion and exclusion. They also addressed UTF-8 encoding issues to ensure proper handling of category names and implemented logging capabilities to track page statistics and save logs to a file. These improvements focused on refining the data extraction process.
wikipedia-dump
zwChan/Wordembedding-and-semantics
Jun 2017 - Apr 2018
Contributions:20 commits, 17 pushes, 2 branches in 9 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.