Julio Perez is a Senior Software Engineer with eight years of experience building GPU-accelerated data and ML infrastructure at NVIDIA, where he was a founding member of the Merlin recommender systems team and designed the first NVTabular feature-engineering pipeline for terabyte-scale workloads. He blends hands-on backend, DevOps and MLOps skills—authoring Dockerfiles, CI/CD pipelines, multi-GPU data loaders and end-to-end deployment flows—to shorten training times and reliably ship containers to public registries. His work touches both TensorFlow and PyTorch integrations and includes visible contributions to high-profile open-source projects like NVIDIA-Merlin/NVTabular and Merlin, demonstrating a focus on performance, compatibility and test stability. A former Army signal intelligence analyst with advanced degrees in cybersecurity and computer science, he brings disciplined systems thinking and operational rigor to complex production ML systems.
8 years of coding experience
5 years of employment as a software developer
Associate of Arts (A.A.), General Studies, 3.3, Associate of Arts (A.A.), General Studies, 3.3 at Broward College
Master of Arts (M.A.), Homeland Security, 3.5, Master of Arts (M.A.), Homeland Security, 3.5 at American Military University
Associate of Science (A.S.), Military Intelligence, General, 4.0, Associate of Science (A.S.), Military Intelligence, General, 4.0 at Cochise College
Bachelor of Science (B.S.), Computer Science and Software Engineering, 3.6, Bachelor of Science (B.S.), Computer Science and Software Engineering, 3.6 at University of Washington Bothell
Bachelor of Arts (B.A.), Political Science and Government, 3.4, Bachelor of Arts (B.A.), Political Science and Government, 3.4 at University of Florida
NVIDIA Merlin is an open source library providing end-to-end GPU-accelerated recommender systems, from feature engineering and preprocessing to training deep learning models and running inference in production.
Role in this project:
DevOps Engineer & MLOps Engineer
Contributions:1 release, 120 reviews, 131 commits in 1 year 5 months
Contributions summary:Julio primarily focused on modifying Dockerfiles and CI/CD configurations, indicating a strong DevOps focus. They made several changes to the build process, including adjusting package installations, updating dependencies, and configuring entry points. Their contributions involved integrating various machine learning and deep learning tools and libraries, revealing MLOps skills related to building and maintaining the project's infrastructure.
NVTabular is a feature engineering and preprocessing library for tabular data designed to quickly and easily manipulate terabyte scale datasets used to train deep learning based recommender systems.
Role in this project:
Back-end & DevOps Engineer
Contributions:311 reviews, 331 commits, 316 PRs in 2 years 8 months
Contributions summary:Julio contributed to the development and management of containerization for the NVTabular project by creating and updating Dockerfiles. They also addressed test failures, indicating a focus on code quality and stability. Further, they modified and updated the code for implementing the integration of various deep learning frameworks, such as PyTorch and TensorFlow.
tsneengineeringtensorflowpreprocessingnvidia
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.