Lu Wang
Software Engineer at Google
California, United States
Summary
🤩
Rockstar🎓
Top SchoolLu Wang is a software engineer at Google with a decade of experience building and modernizing machine learning tooling for mobile and embedded devices. With a PhD in Computer Vision and advanced degrees from top Chinese and US institutions, Lu combines deep research expertise with production-grade engineering—most notably contributing to TensorFlow examples and the widely used TFLite Support toolkit. Their work has modernized image classification reference apps, added GPU acceleration, and expanded metadata and tokenizer support for NLP and audio models, improving deployment and developer ergonomics. Based in California, Lu excels at bridging research and product, turning complex ML models into practical, performant solutions for edge devices.
10 years of coding experience
Doctor of Philosophy (PhD), Computer Vision, Doctor of Philosophy (PhD), Computer Vision at University of Southern CaliforniaMaster's degree, Computer Vision, Master's degree, Computer Vision at Institute of Automation, Chinese Academy of SciencesBachelor's degree, Electronic Engineering, Bachelor's degree, Electronic Engineering at Zhejiang University
Github Skills (16)
javas10
computer-vision10
tflite10
machine-learning10
tokenizer10
tokenize10
metadata10
tensorflow10
java10
image-classification10
natural-language-processing9
android9
c-language8
audio-processing8
gpu-acceleration8
Programming languages (4)
Github contributions (5)
tensorflow/tflite-support
Jul 2020 - Dec 2022
TFLite Support is a toolkit that helps users to develop ML and deploy TFLite models onto mobile / ioT devices.
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 176 reviews, 258 commits in 2 years 5 months
Contributions summary:Lu primarily focused on enhancing the metadata schema for the TensorFlow Lite (TFLite) support repository, specifically for the TFLite model. Their contributions included introducing new features for tokenizers, such as Bert and Regex tokenizers, as well as adding support for audio properties within the content metadata. The user's efforts were geared toward expanding the capabilities of the TFLite metadata system, particularly in support of models employing natural language processing and audio-based models. Furthermore, they modified the testing process to accommodate changes, and they have improved the Java API for image and audio classification models.
iot-devicesmachine-learningtflite-modelstensorflowtflite
tensorflow/examples
Oct 2019 - Dec 2022
TensorFlow examples
Role in this project:ML Engineer
ML EngineerContributions:21 reviews, 23 commits, 1 PR in 3 years 2 months
Contributions summary:Lu primarily focused on migrating and integrating the image classification reference application within the TensorFlow examples repository. Their commits reveal efforts to modernize the application by incorporating the support library and task library. This includes refactoring the code, integrating new APIs, and enabling features like GPU acceleration and improved model support. The user also addressed issues related to switching between delegates in the image classification app.
tensorflow
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.