Masahiro Hiramori
Head Engineer, Open Source Program Office at The Apache Software Foundation
Tokyo, Japan
Summary
🤩
Rockstar🎓
Top SchoolMasahiro Hiramori is a Head Engineer leading Mitsubishi Electric’s Open Source Program Office with 11 years of experience focused on AI compilers, edge inference, and FPGA/ASIC acceleration. He bridges open-source tooling and industrial production, deploying computer vision models and domain-specific LLMs on custom accelerators and shipping user-mode drivers and prototype systems that achieve over 100 FPS on Xilinx UltraScale MPSoC+. An Apache TVM committer and PyTorch Ambassador, he contributes to high-impact projects like TVM and Hummingbird, improving performance, CI/CD, and model frontends. Based in Tokyo, he pairs hands-on compiler and backend engineering with program-level leadership, and maintains developer tools such as a VSCode extension for hardware engineers—evidence of his focus on developer productivity as well as system performance.
11 years of coding experience
8 years of employment as a software developer
Master's degree, Information Science and Technology, Master's degree, Information Science and Technology at Osaka Institute of Technology
English, Japanese
Stackoverflow
91reputation
6kreached
1answer
0questions
Github Skills (31)
pytorch10
verilog10
performance-monitor10
python10
scikit10
machine-learning10
vim-plugin10
neovim10
vim10
compiler-compiler10
performance-analysis10
scikit-learn10
compiler10
tensor10
syntax-checking10
Programming languages (12)
Github contributions (5)
microsoft/hummingbird
Jun 2022 - Jan 2023
Hummingbird compiles trained ML models into tensor computation for faster inference.
Role in this project:
ML Engineer
ML EngineerContributions:8 reviews, 20 commits, 59 PRs in 7 months
Contributions summary:Masahiro primarily contributed to enhancing and maintaining the hummingbird-ml library. They focused on updating dependencies such as TVM and Prophet, and fixed deprecation warnings for sklearn and scipy, ensuring compatibility and improving code quality. The contributions also include adding support for Bayesian Gaussian Mixture Models, indicating the user's involvement in expanding the library's machine learning capabilities. Further commits involved fixing shape errors in numpy array creation.
pytorchcomputationml-modelspythondeep-learning
apache/tvm
Feb 2021 - Nov 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:30 reviews, 11 commits, 96 PRs in 1 year 9 months
Contributions summary:Masahiro primarily contributed to performance optimization and bug fixing within the TVM compiler stack. Their work involved addressing minor misspellings and fixing inconsistencies in code related to batch matrix multiplication and other tensor operations. They also made changes in the relay and nnvm modules. The user demonstrates proficiency in compilers and machine learning operations.
metalvulkancompilertensoropencl
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.