Jiren Jin
Senior Software Engineer at Google
Tokyo, Japan
Summary
👤
Senior🎓
Top SchoolJiren Jin is a Senior Software Engineer at Google in Tokyo with 11 years of experience building production ML and serving systems for large-scale products like Google Translate and Search Verticals. He has designed data pipelines, metrics systems, and APIs for LLM-based features, and most recently created a framework to post-correct GenAI outputs for product-grade reliability. Trained in AI and machine learning through doctoral studies at The University of Tokyo, he blends research rigor—experience in continual learning and domain adaptation—with production engineering. His open-source contributions include back-end and training-loop improvements to the well-known Chainer deep-learning framework, where he implemented dynamic training extensions and fixed optimizer/quality issues. Comfortable across research and production, he brings a rare mix of deep ML expertise and hands-on systems engineering for deployed language and search products.
11 years of coding experience
3 years of employment as a software developer
University of TokyoBachelor of Science (BS), Information Engineering, Bachelor of Science (BS), Information Engineering at Beijing University of Posts and Telecommunications
Chinese, English, Japanese
Stackoverflow
377reputation
65kreached
6answers
0questions
Github Skills (18)
python10
artificial-neural-networks10
chainer10
machine-learning10
deeplearning-ai10
deep-learning10
neural-network10
numpy9
documentation7
cuda7
cmd6
tensorflow-datasets6
tensorflow6
pylint6
ipython6
Programming languages (11)
Github contributions (5)
chainer/chainer
Aug 2017 - Dec 2018
A flexible framework of neural networks for deep learning
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:14 commits, 9 PRs, 21 comments in 1 year 4 months
Contributions summary:Jiren primarily contributed to the Chainer deep learning framework by addressing bugs related to optimizer behavior and freezing layers. They implemented the `InverseShift` extension for dynamic attribute adjustments during training, demonstrating familiarity with training loop mechanics. Further contributions involved fixing flake8 errors, adding a `StepShift` extension for training rate control, and integrating the `InverseShift` extension in the documentation, indicating a focus on framework maintenance and enhancement.
deep-learningneural-networkpythonneural-networksmachine-learning
JirenJin/recurrent-image-annotator-web-demo
Jul 2016 - Mar 2019
Contributions:114 commits, 58 pushes, 3 branches in 2 years 8 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.