Qianli Zhu
Senior Software Engineer at Google
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolQianli Zhu is a Senior Software Engineer at Google with eight years focused on TensorFlow, Keras, and ML tooling, based in Mountain View. He bridges research and production by contributing to high-impact open-source projects—improving core Keras RNNs and CV workflows in keras-cv, extending the XLA compiler, and enabling DTensor-backed distributed training in TensorFlow. Comfortable across backend systems and model-level code, he has refactored APIs, implemented augmentation and segmentation features, and helped bring model optimization (quantization/pruning) into production-ready form. His background spans Java/J2EE and iOS consulting to deep learning infrastructure, giving him a pragmatic view of both engineering practices and ML research. Colleagues rely on him for clean refactors that preserve compatibility while advancing functionality, evidenced by migrations across tensorflow/addons, estimator, and google-research.
8 years of coding experience
3 years of employment as a software developer
Master of Information Technology, Database, Network, Master of Information Technology, Database, Network at University of New South WalesBachelor of Engineering, Electrical Engineering, Bachelor of Engineering, Electrical Engineering at Shanghai University
English, Chinese, Japanese
Github Skills (57)
algorithm10
algorithms10
benchmark10
c-language10
distributed-training10
bigquery10
ai-agent10
python10
apidoc10
machine-learning-models10
machine-learning10
benchmarking10
rnn-model10
n10
reinforcement-learning10
Programming languages (6)
Github contributions (5)
tensorflow/estimator
Mar 2019 - Nov 2021
TensorFlow Estimator
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 review, 28 commits, 1 comment in 2 years 8 months
Contributions summary:Qianli primarily focused on updating and refactoring the TensorFlow Estimator library to utilize the latest Keras API and address performance issues related to RNN implementations. They updated the code to use Keras RNN V2 layers, improving performance in multiworker setups. Additionally, the user removed private Keras usages and ensured that the Estimator utilized public APIs, streamlining the transition to the OSS Keras backend. These changes reflect a focus on modernizing the code while maintaining its functionality.
deep-learningmachine-learningtensorflowtensorflow-estimatorestimator
keras-team/keras-cv
Oct 2021 - Oct 2022
Industry-strength Computer Vision workflows with Keras
Role in this project:ML Engineer
ML EngineerContributions:340 reviews, 25 commits, 52 PRs in 1 year
Contributions summary:Qianli primarily worked on enhancing the `keras-cv` library, focusing on computer vision tasks. Their contributions centered around improving bounding box utilities, including readability and functionality. They implemented and tested preprocessing layers for image augmentation, like random saturation, hue, and sharpness. The user also added a segmentation head and developed the foundation for a DeepLabV3 model.
visiondeep-learningcomputer-visionindustrystrength
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.