Uday Bondhugula
CTO at PolyMage Labs
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolUday Bondhugula is a CTO and professor with 15 years of experience at the intersection of compilers, high-performance computing, and AI accelerators, co-founding the influential MLIR project within the LLVM ecosystem. He has repeatedly founded and led PolyMage Labs, translating research in automatic parallelization and the polyhedral framework into practical compiler technology and startup productization. His research-grade contributions to MLIR, TensorFlow MLIR, LLVM, and Halide focus on affine loop optimizations, loop fusion/tiling, and performance-driven transformations that bridge academic ideas and production toolchains. Based in Bengaluru and holding a PhD from Ohio State, he combines academic leadership at IISc with hands-on backend engineering on widely used open-source compiler projects. A less obvious strength is his knack for turning sophisticated program analysis into measurable performance gains—evident from benchmarking and auto-scheduling work in Halide alongside MLIR improvements.
15 years of coding experience
16 years of employment as a software developer
Indian Institute of Technology MadrasPh.D., Computer Science and Engineering, Ph.D., Computer Science and Engineering at The Ohio State UniversityHigh School, High School at Little Flower Junior College, Uppal
English, Hindi, Telugu, French, Kannada
Github Skills (19)
benchmark10
performance-monitor10
performance-analytics10
c-language10
halide10
benchmarking10
mlr10
performance-measurement10
compiler-compiler10
performance-analysis10
affine-transformation10
compiler-optimization10
performance-tuning10
compiler10
performance-monitoring10
Programming languages (14)
Github contributions (5)
tensorflow/mlir
Nov 2019 - Dec 2019
"Multi-Level Intermediate Representation" Compiler Infrastructure
Role in this project:
Back-end Developer
Back-end DeveloperContributions:10 commits, 47 PRs, 334 comments in 1 month
Contributions summary:Uday primarily contributed to the MLIR compiler infrastructure, focusing on improving the `isValidSymbol` function and related code. They refactored code to enhance its power and accuracy for dimension and symbol checking. The user also addressed issues by implementing optimizations for dimension folding and introduced prefetch operations to improve performance. These changes indicate a focus on compiler optimization and the underlying infrastructure for high-performance computing.
representationmulti-levelinfrastructurelevel-intermediatecompiler
llvm/llvm-project
Jan 2023 - Jan 2023
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:Back-end Developer
Back-end DeveloperContributions:80 reviews, 1 commit, 110 PRs in 1 day
Contributions summary:Uday primarily contributed to the MLIR (Multi-Level Intermediate Representation) project, focusing on the Affine dialect. Their contributions included fixing bugs, refactoring code for efficiency, and improving the analysis and transformation utilities related to affine loops. These changes involved improvements to the loop fusion and loop tiling passes. Their work aimed at enhancing the performance and capabilities of the compiler infrastructure.
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.