Théo Degioanni
Deep Learning Compiler Engineer at NVIDIA
Greater Paris Metropolitan Region France
Summary
🤩
Rockstar🎓
Top SchoolThéo Degioanni is a Deep Learning Compiler Engineer with eight years of experience bridging compiler infrastructure, MLIR/CIRCT hardware compilation, and performance-focused toolchain work. Now at NVIDIA working on the TileIR spec and compiler, he has a strong research-to-production trajectory from internships at Cambridge, NextSilicon, Edinburgh and Inria where he designed MLIR-based passes, hardware simulation backends and analyses for Mem2Reg/SROA. A consistent open-source contributor, he has improved LLVM’s MLIR support (i0 canonicalization, C API dialect loading) and contributed rendering and tutorial fixes to the Rust game engine Amethyst, showing fluency across low-level compiler internals and practical systems. He combines formal dialect/meta-dialect design with pragmatic testing and error-message improvements, and often surfaces non-obvious correctness and compatibility fixes that enable downstream tooling.
8 years of coding experience
Classe préparatoire aux grandes écoles, Mathematics and Physics, Classe préparatoire aux grandes écoles, Mathematics and Physics at Lycée Michelet, VanvesMaster's Degree, Computer Science, Master's Degree, Computer Science at École normale supérieure de Rennes
Github Skills (13)
c-language10
rust10
engine10
game-development10
compiler-construction10
game-engine10
cprogramming-language10
mlr10
render10
game-dev10
llvm10
sprite10
testing9
Programming languages (14)
Github contributions (5)
amethyst/amethyst
Apr 2018 - May 2019
Data-oriented and data-driven game engine written in Rust
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:40 commits, 45 PRs, 9 pushes in 1 year 1 month
Contributions summary:Théo primarily focused on adding sprite rendering utilities and improving the Pong tutorial examples. They implemented features for sprite rendering, including the creation of sprite renderers and related components. Furthermore, the user updated and refactored the Pong tutorial by fixing broken links, correcting grammar, and fixing mistakes in the tutorial code. They also updated examples and fixed a merge issue in the window example.
data-drivengamegamedevrustentity-component
llvm/llvm-project
Mar 2023 - Apr 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:Back-end Developer
Back-end DeveloperContributions:39 reviews, 20 PRs, 15 pushes in 2 years
Contributions summary:Théo primarily contributed to the LLVM project by adding support for zero-width integers in MathExtras.h, addressing compatibility issues with MLIR. They also fixed a typo in a unit test and added tests for i0 canonicalization in the MLIR/arith dialect. Additional contributions included adding IRDL dialect loading to the C API and fixing related bugs and improving error messages. These commits showcase a focus on compiler infrastructure, testing, and improving language toolchain functionality.
compilerstechnologiesclangsubmittoolchain
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.