Terry Heo
Software Engineer at Google
Palo Alto, California, United States
Summary
🤩
Rockstar🎓
Top SchoolTerry Heo is a seasoned software engineer with 15 years of experience focused on on-device ML and systems software, currently contributing to TensorFlow Lite at Google from Palo Alto. He brings deep expertise in backend kernel work, performance profiling, and build automation—evidenced by contributions that integrate XNNPACK transpose into TFLite kernels and add ARM wheel builds and manylinux2014 support for TensorFlow Lite. His background spans embedded/IoT and Android projects as well as VoIP systems from earlier roles at LG and LG-Nortel, giving him a pragmatic systems-first perspective. Colleagues rely on him for solving tricky cross-platform linking and build pipeline issues that often block releases, and he pairs that with a research-rooted view from an MS in Computer Engineering.
15 years of coding experience
11 years of employment as a software developer
Master, Computer Engineering, Master, Computer Engineering at Pusan National University
Stackoverflow
159reputation
19kreached
20answers
0questions
Github Skills (28)
c-language10
python10
architecture10
machine-learning10
bash10
cicd10
ubuntu10
tensorflow-lite10
tensorflow10
performance-optimization10
bazel10
build-automation10
arm10
cprogramming-language10
architectures10
Programming languages (5)
Github contributions (5)
tensorflow/tensorflow
Sep 2019 - Jan 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:230 reviews, 362 commits, 45 PRs in 3 years 5 months
Contributions summary:Terry contributed to the TensorFlow Lite kernel, specifically calling XNNPACK Transpose from the TFLite kernel. They also updated a tool related to delegate performance benchmarking by allowing the stable ABI delegate provider to load JSON-formatted delegate settings. Additionally, the user implemented code to capture heap allocations in the Xprof profiler for the purpose of performance profiling. The commits demonstrate expertise in optimizing and integrating components within the TensorFlow ecosystem, including kernel implementation and delegate settings.
pythondata-sciencedeep-learningmlmachine-learning
tensorflow/tflite-support
Nov 2021 - Apr 2022
TFLite Support is a toolkit that helps users to develop ML and deploy TFLite models onto mobile / ioT devices.
Role in this project:DevOps Engineer & Automation Engineer
DevOps Engineer & Automation EngineerContributions:1 review, 7 commits, 2 PRs in 5 months
Contributions summary:Terry primarily focused on automating the build process and expanding the support for building TensorFlow Lite wheels for ARM architectures. This involved creating scripts for building Python wheels for ARM systems, integrating Ubuntu 18 support, and updating the build process for release configurations. The user also worked on integrating licensing and updating the build platform to `manylinux2014`. Furthermore, the user addressed a linking issue on ARM platforms.
iot-devicesmachine-learningtflite-modelstensorflowtflite
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.