Ian Chang
Software Technical Lead at Google
Kirkland, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolIan Chang is a Software Technical Lead with seven years of focused experience designing power-, thermal-, and performance-aware system software for mobile SoCs and edge devices. At Google he leads cellular/modem power and thermal software, defines multi-year roadmaps, and drives AI-enabled user-centric optimizations across demanding journeys like video calls and Gemini, earning multiple internal awards. He combines deep platform expertise from roles at OnePlus, Intel, and HTC—holding numerous patents—with hands-on work in secure ML inference and private compute prototypes. As a contributor to projects like project-oak’s TFLite integration, he bridges MLOps and full-stack engineering to validate models and support multi-tensor workflows. Ian also founded RoboTradee, applying agentic financial LLMs to automated trading, showing a penchant for turning research-grade AI into product-facing systems. Based in Kirkland, WA, he blends system-level rigor with product vision to squeeze the best UX out of constrained hardware.
7 years of coding experience
9 years of employment as a software developer
Master's degree Computer Science and Information Engineering, Master's degree Computer Science and Information Engineering at National Chung Cheng University
Github Skills (10)
tensorflow-lite10
c-language10
cprogramming-language10
bazel10
build-automation9
machine-learning9
distributed-systems9
python8
policies8
enclave7
Programming languages (5)
Github contributions (5)
project-oak/oak
Oct 2022 - Nov 2022
Meaningful control of data in distributed systems.
Role in this project:
Full-stack Developer & MLOps Engineer
Full-stack Developer & MLOps EngineerContributions:43 reviews, 7 commits, 17 PRs in 29 days
Contributions summary:Ian contributed significantly to the project's TensorFlow Lite integration, including the development of a generic TFLite model runner and the implementation of a hello_world model. They added support for multi-tensor inputs and outputs. Furthermore, the user worked on supporting the REDUCE_ANY OP within the TFLite framework, improving the flexibility of the project. Their work also involved setting up the build process and validating model accuracy.
policydistributed-systemsdatabaseenclavemeaningful
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.