Giorgio Patrini
Co-Founder, CEO And Chief Scientist at Sensity
Amsterdam, North Holland, Netherlands
Summary
🤩
Rockstar🎓
Top SchoolGiorgio Patrini is a Co-Founder, CEO and Chief Scientist based in Amsterdam with 11 years of experience building Sensity, a company that pioneered enterprise-grade detection of AI-generated and digitally altered media. Under his technical leadership Sensity protects clients across four continents in finance, marketplaces, insurance and government and is backed by investors including Betaworks, Amadeus Capital and the European Commission. He holds a PhD in Machine Learning and his research spans deep generative models, federated learning and homomorphic encryption, bringing privacy-aware rigor to product design. An active open-source contributor to scikit-learn, he has improved PCA numerical stability, implemented partial_fit for scalers and refined randomized SVD—work that reveals uncommon depth in both ML theory and production engineering. He also advises startups and has helped founders navigate acquisitions, bridging academic research with commercial impact.
11 years of coding experience
2 years of employment as a software developer
Australian National University
Italian, English, French
Github Skills (15)
scikit10
scikit-learn10
machine-learning10
pca10
python10
data-science10
numpy10
data-analysis10
statistics9
data-structure7
datastructures7
datastructure7
data-structures7
algorithm7
algorithms7
Programming languages (8)
Github contributions (5)
scikit-learn/scikit-learn
Aug 2015 - Mar 2016
scikit-learn: machine learning in Python
Role in this project:
Data Scientist
Data ScientistContributions:24 commits, 22 PRs, 488 comments in 7 months
Contributions summary:Giorgio primarily contributed to improving the scikit-learn library by addressing numerical stability issues and enhancing existing functionalities. This involved removing potential runtime warnings in the PCA module, specifically in a corner case of the fit method. The user also implemented the partial_fit method for various scalers, extending their usability for online learning scenarios. Furthermore, the user introduced and documented changes related to the randomized SVD implementation, including improvements to the power iteration normalizer.
data-analysispythonstatisticsdata-sciencelearn-machine-learning
giorgiop/scikit-learn
Aug 2015 - Oct 2016
scikit-learn: machine learning in Python
Contributions:334 pushes, 41 branches in 1 year 2 months
pythondata-sciencelearn-machine-learningneural-networksmachine-learning
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.