Robert Suderman
Staff Software Engineer at AMD
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolRobert Suderman is a Staff Software Engineer in Seattle with 11 years of experience fusing compiler engineering and machine learning to productionize efficient models and runtimes. He was a founding contributor and TL on Google’s IREE project, driving front-end compilation, TOSA transformations, and quantized/mixed-precision support across TensorFlow, JAX, and TFLite backends. At AMD he continues to bridge ML research with compiler tech, and his open-source work shows hands-on depth from 360° video tooling and GUI work to low-level MLIR/retargetable compilation and quantized inference optimizations. Comfortable across Python, C++ and MLIR toolchains, he’s equally likely to ship a Tkinter utility as to implement quantized multi-head attention and lowering passes. Collected experience from AR/VR playback at YouTube to compiler internals gives him a rare blend of systems, multimedia, and ML deployment expertise.
11 years of coding experience
12 years of employment as a software developer
Bachelor of Mathematics Computer Science, Bachelor of Mathematics Computer Science at University of Waterloo
Github Skills (44)
tosca10
stable10
python10
attention-mechanism10
machine-learning10
lsa10
mlr10
togaf10
deep-learning10
tensorflow10
tkinter10
quantization10
irr10
tflite10
nlp10
Programming languages (6)
Github contributions (5)
google/spatial-media
Mar 2015 - Nov 2017
Specifications and tools for 360º video and spatial audio.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:41 commits in 2 years 8 months
Contributions summary:Robert primarily contributed to the development of a Python-based tool for processing and injecting metadata into video files, specifically for 360-degree videos. Their work included refactoring the code, improving extension parsing for different video formats (.mp4, .mov), and fixing indentation issues. Furthermore, the user implemented the graphical user interface (GUI) using Tkinter, and made improvements for GUI-related behaviors like file saving.
spatialaudiospatial-audiospecificationsvideo
iree-org/iree
Sep 2019 - Jan 2023
A retargetable MLIR-based machine learning compiler and runtime toolkit.
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:520 reviews, 314 commits, 930 PRs in 3 years 4 months
Contributions summary:Robert primarily contributed to the development of machine learning model-related functionalities within the IREE compiler and runtime toolkit. They implemented the decomposition of several mhlo operations and added the lowering of those and other operations to run within IREE execution. They also added various tests to verify those operations, and contributed to the overall support for the TOSA intermediate representation. The focus of the user's work involved enhancing the system's ability to handle quantized and mixed-precision operations.
mlirspirvvulkantensorflowcompiler
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.