Matt Welsh
Engineering Manager at OctoML
Seattle, Washington, United States
Summary
🤩
RockstarMatt Welsh is an engineering manager and VP of Engineering with a focused background in embedded systems and edge AI, based in Seattle. Over five years of experience, he combines hands-on firmware and IoT engineering with team leadership to deliver production-ready edge deployments. He has notable open-source contributions to the widely used Apache TVM project, where he brought up support for the nRF5340DK and improved the Zephyr runtime and QEMU test integrations to advance microTVM on constrained devices. Comfortable navigating build systems, emulators, and board bring-up, Matt blends low-level debugging chops with pragmatic product delivery. Colleagues rely on him to translate tricky hardware-software constraints into reliable tooling and repeatable test workflows.
5 years of coding experience
Github Skills (10)
c1710
embedded10
c1110
sys10
zephyr10
qemu9
testing9
compiler-compiler8
machine-learning8
compiler8
Programming languages (4)
Github contributions (5)
apache/tvm
Jan 2021 - Aug 2021
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
Embedded Systems Engineer / IoT Developer
Embedded Systems Engineer / IoT DeveloperContributions:32 reviews, 10 commits, 10 PRs in 7 months
Contributions summary:Matt contributed significantly to the microTVM project, particularly focusing on bringing up support for the nRF5340DK development board. Their work involved modifying existing code, fixing build issues, and integrating the nRF5340DK into the test suite, including the QEMU emulator, aiming to enable edge AI deployment. The commits also feature improvements and enhancements for the Zephyr demo runtime, including corrections, and updates for improved model execution, and debugging efforts, suggesting a deep understanding of the embedded systems environment. Furthermore, they addressed various issues related to compilation and flashing, involving the use of `west` command, and the introduction of additional features for the Zephyr runtime.
metalvulkancompilertensoropencl
mdw-octoml/tvm
Jan 2021 - Aug 2021
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:93 commits, 55 pushes, 12 branches in 7 months
cpugpu-programminggpu-accelerationtvmdeep-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.