Deqiang Chen
Software Engineer at Google
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolDeqiang Chen is a software engineer in San Francisco with 7 years focused on systems-level machine learning runtimes and a long background in embedded and firmware engineering. Currently at Google, he builds high-performance ML runtime infrastructure—contributing to TensorFlow and TFRT integration work—and has a track record of making ML deployment faster and more efficient for edge devices. Prior roles at Qualcomm and Texas Instruments honed his low-level C/C++ expertise across 3G/4G/5G modem firmware and audio codecs, giving him deep experience in performance-critical systems. He also contributed to TensorFlow Lite Micro, improving operator coverage, kernel performance and test reliability for constrained devices. Comfortable across C++, Python, Java and compiler tech like MLIR, he combines research-level rigor (PhD in EEE) with pragmatic production delivery. An engineer who moves between firmware and cloud ML runtimes, he often bridges hardware constraints and scalable software design in shipping complex systems.
7 years of coding experience
10 years of employment as a software developer
University of Science and Technology of ChinaMaster's degree, Electrical Engineering, Master's degree, Electrical Engineering at University of Notre Dame
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (25)
c-language10
testing10
configuration-management10
machine-learning10
tensorflow-lite10
mlops10
tensorflow10
performance-optimization10
sys10
embedded10
cpp10
cprogramming-language10
protocol-buffers9
gnu-make9
continuous-integration9
Programming languages (3)
Github contributions (5)
tensorflow/tflite-micro
Aug 2021 - Jul 2022
Infrastructure to enable deployment of ML models to low-power resource-constrained embedded targets (including microcontrollers and digital signal processors).
Role in this project:
ML Engineer
ML EngineerContributions:358 reviews, 101 commits, 144 PRs in 11 months
Contributions summary:Deqiang primarily contributed to the development and testing of the TensorFlow Lite Micro framework, specifically focusing on improving operator coverage and optimizing kernel performance. Their work included adding new test cases for the convolution operator, enhancing data precision, and refactoring build processes. The user also contributed to the integration of new tests within the nightly build test suite and improving the reliability of existing tests by fixing platform-specific issues, demonstrating a focus on improving the quality and scope of the testing infrastructure for the embedded machine learning models. The user also worked on tools and documentation that relates to code and memory footprint.
signalml-modelslow-powerprocessorsdeployment
tensorflow/tensorflow
Mar 2022 - Apr 2022
An Open Source Machine Learning Framework for Everyone
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:2 commits in 1 month
Contributions summary:Deqiang contributed to the TensorFlow project with commits focusing on improving the integration of the TFRT runtime with the MLRT context and saved model features. The user implemented changes to the graph executor, including connecting cancellation contexts and adding passes to the compiler to assign op keys. They also worked on enabling and testing end-to-end cancel functionality for the MLRT runtime and refactoring pipelines to allow for flexible reuse and enabling while parallel iterations.
pythondata-sciencedeep-learningmlmachine-learning
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.