Davide Testuggine
Software Engineer at Meta
Redwood City, California, United States
Summary
🤩
Rockstar🎓
Top SchoolDavide Testuggine is an AI-focused software engineer with 13 years of experience building privacy-preserving and multi-modal ML systems at scale, currently at Meta after leadership roles at Snap and Facebook. He led development of Opacus and helped grow the differential privacy ecosystem—contributing backend integration, secure RNG support, and automation to a high-profile PyTorch project used across industry. At Snap he served as tech lead for content moderation and understanding, delivering NLP, CV, and graph-learning platforms for large-scale ranking and safety. His background spans product-facing ML at Microsoft (Bing Predicts and core relevance) and academic work in security and NLP, giving him a rare blend of research rigor and production engineering. Based in Redwood City, he combines platform-level thinking with hands-on code and a track record of evangelizing PPML through talks and community collaborations.
13 years of coding experience
9 years of employment as a software developer
Master of Philosophy (MPhil), Advanced Computer Science, Master of Philosophy (MPhil), Advanced Computer Science at University of CambridgeLaurea, Information Engineering, Laurea, Information Engineering at Università di Genova
italian (native), English, French
Github Skills (8)
pytorch10
machine-learning10
python10
deeplearning-ai9
deep-learning9
architectures7
architecture7
git6
Programming languages (10)
Github contributions (5)
meta-pytorch/opacus
Dec 2019 - Jan 2023
Training PyTorch models with differential privacy
Role in this project:
Backend Developer
Backend DeveloperContributions:2 releases, 19 reviews, 44 commits in 3 years 2 months
Contributions summary:Davide primarily contributed to the `pytorch/opacus` repository by modifying code related to training PyTorch models with differential privacy. Their work involved refactoring and deleting unnecessary files, adding a website, and integrating the project with `torchcsprng` for secure random number generation. Furthermore, the user automated the code formatting process and addressed website-related issues.
differential-privacypytorchprivacy-preserving-machine-learningdeep-learningmachine-learning
Darktex/opacus
Aug 2020 - Jan 2023
Training PyTorch models with differential privacy
Contributions:100 pushes, 35 branches in 2 years 5 months
pytorchpytorch-modelsprivacydeep-learningdifferential-privacy
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.