Fergus Henderson
Staff Software Engineer at Google
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolFergus Henderson is a Staff Software Engineer based in London with 18 years of experience designing and shipping robust ML and systems software, currently leading advanced work at Google. He combines deep expertise in TensorFlow/TFLite and MediaPipe—contributing to GPU acceleration, custom operator support, and stable C API integrations—with practical systems engineering skills in build and release tooling (distcc). His work spans low-level C/C++ architecture, memory and delegate management for edge ML, and improving portability and test coverage, reflecting a rare blend of research-era rigor and production-first pragmatism. Known for quietly strengthening foundations—refactors that enable binary compatibility and clearer build processes—he improves both developer experience and runtime reliability.
18 years of coding experience
4 years of employment as a software developer
University High School acceleration program, Victoria, AustraliaThe University of Melbourne
Stackoverflow
86reputation
8kreached
2answers
0questions
Github Skills (38)
application-framework10
c-language10
api-design10
app-framework10
machine-learning10
build-system10
capi10
makefile10
automation10
tensorflow-lite10
delegates10
tensorflow10
cplus10
web-framework10
automations10
Programming languages (7)
Github contributions (5)
distcc/distcc
Mar 2008 - Jun 2022
distributed builds for C, C++ and Objective C
Role in this project:
Back-end Developer & Build & Release Engineer
Back-end Developer & Build & Release EngineerContributions:1 review, 202 commits, 1 PR in 14 years 5 months
Contributions summary:Fergus primarily contributed to improving the build and installation process for distcc. Their work included fixing portability issues, adapting the build system to support running autogen.sh from the build directory, and modifying configuration and installation files. A significant portion of the commits focused on refining the build process, including modifications to packaging scripts (deb.sh, rpm.sh) and the Makefile, as well as adding build-related features like '--scan-includes' and handling the --sysroot option. Further changes involved modifying the code to address compiler warnings.
objective-ccppinfinibandbuildsc-plus-plus
tensorflow/tflite-support
Aug 2020 - Jan 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:5 reviews, 58 commits, 11 comments in 1 year 5 months
Contributions summary:Fergus primarily contributed to the TensorFlow Lite Support library, focusing on the core engine and vision task implementations. They refactored code to utilize the TF Lite shims, enabling the use of the TF Lite stable API. Key contributions include modifying the codebase to support the TF Lite C API, enhancing the interpreter creation process, and updating dependencies to leverage recent TensorFlow snapshots. These changes involved updating core functionalities, such as improving error handling and streamlining the use of the TF Lite API within the context of machine learning model deployment to mobile and IoT devices.
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.