Tori Baker - Software Engineer at Google

Tori Baker

Software Engineer at Google

Munich, Bavaria, Germany

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Tori Baker is a software engineer with 11 years of experience, currently focused on code generation for ML compilers targeting GPUs at Google in Munich. She has deep backend and performance engineering expertise, contributing to high-profile open-source projects like Triton (GPU compiler optimizations) and Orbit (C/C++ profiler), where she improved stability, performance calculations, and UI data handling. Her background spans systems engineering for large-scale platforms—from VM image tooling at Stadia to build/test reliability for the IREE ML compiler—and includes practical experience in Go, C/C++, Python, and GPU toolchains. Tori’s work emphasizes correctness under hardware constraints (e.g., shared memory, warp layouts, division-by-zero handling) and improving test and presubmit infrastructure to prevent regressions. She combines research-minded debugging with production-focused automation, and often surfaces non-obvious fixes that improve both compiler IR generation and runtime behaviour.

11 years of coding experience

6 years of employment as a software developer

California Polytechnic State University, San Luis Obispo

Github Skills (40)

performance-monitor10

performance-analytics10

c-language10

back-end-development10

python10

code-profiling10

llvm10

testing10

matrix-multiplication10

gpu-programming10

mlr10

triton10

compiler-compiler10

performance-analysis10

profiling10

Programming languages (2)

C++LLVM

Github contributions (5)

google/orbit

Jul 2021 - Jan 2023

C/C++ Performance Profiler

Role in this project:

Backend Developer & Performance Engineer

Contributions:184 reviews, 49 commits, 99 PRs in 1 year 6 months

Contributions summary:Tori made significant contributions to the Orbit performance profiler, specifically adding features related to standard deviation calculations and incorporating them into the live data views. They also optimized the UI by distinguishing the "All Threads" event track and adjusting its height and tooltip. Furthermore, the user refactored the standard deviation calculations to ensure accuracy by recalculating them upon capture completion. This indicates a focus on data processing, performance analysis, and UI enhancements.

instrumentationhookingvisualizeretwperformance

triton-lang/triton

Jul 2023 - Mar 2025

Development repository for the Triton language and compiler

Role in this project:

Backend Developer

Contributions:14 reviews, 49 PRs, 43 pushes in 1 year 8 months

Contributions summary:Tori primarily contributed to the backend aspects of the Triton language and compiler, focusing on optimizations and bug fixes within the TritonGPU backend. Their work involved improving code related to shared memory access, layout conversions, and handling potential division-by-zero scenarios. Key contributions include fixing issues related to incorrect warp order calculations for MMA_V3 layouts and improving the stability of the compiler. The user also added and modified tests related to the LLVM IR generation and the Triton compiler's behavior.

compilerprogramming-languagecode-generationtriton

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.

Request Free Trial