Leo Blonk
Principal Engineer at ARM
Ireland
Summary
🤩
RockstarLeo Blonk is a Principal Engineer based in Ireland with over 30 years of software engineering experience and six years focused on machine learning compiler development. At ARM since 2014 he blends deep systems and embedded software expertise with compiler work, contributing to TVM by improving LLDB pretty-printers for Relay, adding operator support like Tanh, and enabling per-channel quantization and hardware-specific integrations for Ethos-N. His background spans product and project leadership across SoC design tools, image processing, and travel software, giving him a rare combination of hands-on low-level engineering and product delivery skills. Known for making complex data structures more debuggable, he brings a pragmatic focus on tooling that accelerates ML deployment on specialized accelerators.
6 years of coding experience
25 years of employment as a software developer
Github Skills (11)
compiler10
debugging10
tvm10
machine-learning10
deeplearning-ai10
compiler-compiler10
deep-learning10
lldb10
cprogramming-language9
c-language9
python8
Programming languages (3)
Github contributions (5)
apache/tvm
Dec 2019 - May 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:24 reviews, 23 commits, 24 PRs in 2 years 5 months
Contributions summary:Leo's primary contribution involves developing and integrating pretty printers for the LLDB debugger, specifically for Relay data structures within the TVM project. This enhancement facilitates debugging by enabling the display of complex data structures. Furthermore, the user contributed to the Ethos-N integration, by updating test hashes and integrating support for non-default configurations. These modifications indicate a focus on improving debugging tools and supporting specific hardware configurations within the deep learning compiler stack. Finally, the user implemented new support and expanded existing support for new operators such as Tanh and per-channel quantization.
metalvulkancompilertensoropencl
Leo-arm/tvm
Nov 2019 - May 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:59 pushes, 45 branches in 2 years 5 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.