Guilherme Perrotta

Research Engineer at Google DeepMind

São Paulo, Brazil
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
🎓
Top School
Guilherme Perrotta is a research engineer based in São Paulo with eight years of experience at the intersection of systems, compiler research, and applied machine learning. He currently researches compilers for ML at Google DeepMind after roles at Google Research where he helped deliver state-of-the-art image matching models and built experiment infrastructure in PyTorch and Apache Beam. His background spans distributed runtimes, CUDA/OpenMP optimizations, and production NLP/ backend work—skills honed in research internships at Petrobras and Samsung and in product-focused development at NeuralMind. A master's graduate from Universidade Estadual de Campinas with top grades, he combines rigorous academic training with hands-on engineering that moves models from experiments into internal libraries and services. Notably, he has repeatedly bridged low-level performance engineering and high-level ML modelling, enabling both faster runtimes and reproducible research workflows.
code8 years of coding experience
job7 years of employment as a software developer
bookMaster's degree, Computer Science, GPA 4 of 4, Master's degree, Computer Science, GPA 4 of 4 at Universidade Estadual de Campinas
languagesPortuguese, English
github-logo-circle

Github Skills (19)

python9
rio9
linux9
windows9
projeto9
configurations9
scientific-computing9
macos9
package-manager8
hpc8
multiple-versions8
build-tools7
spack5
deep-learning5
infrastructure5

Programming languages (2)

C++Python

Github contributions (5)

github-logo-circle
gperrotta/devito

Nov 2019 - Nov 2019

Code generation framework for automated finite difference computation
Contributions:3 pushes in 6 days
computationcode-generationpythontlafinite-difference
gperrotta/yask

Oct 2019 - Nov 2019

YASK--Yet Another Stencil Kit: a domain-specific language and framework to create high-performance stencil code for implementing finite-difference methods and similar applications.
Contributions:4 pushes in 10 days
similarfinite-differencefinite-difference-methodsperformancestencil
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
Guilherme Perrotta - Research Engineer at Google DeepMind