Diego Gomez is a Senior Software Engineer and Machine Learning specialist with 11 years of experience, currently building large-scale image, video and text safety classifiers for YouTube at Google from Paris. He designs and maintains ML infrastructure to deploy production models at scale and leads intern mentorship programs, blending hands-on engineering with people development. His open-source contributions include extending core tensor operations in PySyft for privacy-preserving data science and implementing foundational GAN models in PyTorch/TensorFlow, reflecting both applied ML and systems-level expertise. Diego’s background includes published research in astronomical transient recognition and a Cum Laude Computer Science degree from Universidad de los Andes, showing strong academic rigor. Comfortable across back-end systems, model training pipelines and research code, he brings a rare mix of production impact and research-driven curiosity.
11 years of coding experience
5 years of employment as a software developer
Universidad de los Andes
Nanodegree, Machine Learning Engineer, Nanodegree, Machine Learning Engineer at Udacity
Generative Adversarial Networks implemented in PyTorch and Tensorflow
Role in this project:
ML Engineer
Contributions:71 commits, 7 PRs, 57 pushes in 3 years 6 months
Contributions summary:Diego contributed to the implementation of Generative Adversarial Networks (GANs) within the PyTorch and TensorFlow frameworks. Their commits demonstrate the creation of basic MNIST loading, model definition (discriminator and generator networks), and initial training routines. The user's work focused on setting up the core components of a Vanilla GAN architecture, indicating an emphasis on foundational ML concepts.
Perform data science on data that remains in someone else's server
Role in this project:
Back-end Developer
Contributions:9 commits, 4 PRs, 20 comments in 19 days
Contributions summary:Diego primarily contributed to the core functionality of the `pysyft` library by adding several tensor operations. These include functions like `neg`, `triu`, `ceil`, `floor`, and `sigmoid`, demonstrating a focus on expanding the library's capabilities for tensor manipulation. Additionally, the user appears to be involved in updating the library's build and installation configuration.
pytorchcryptographyacquiringpythonscience
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.