Open Source

I have (co-)implemented and released three Python packages for (Differentially Private) synthetic data generation. They are all beginner friendly and have been used in academia (UCL, University of Oxford, Imperial College London, The Alan Turing Institute) and industry (SAS, HSBC, Oblivious) settings:

• dpmm – Differentially Private Marginal Models, a Library for Synthetic Tabular Data Generation (published at TPDP 2025)
• dpart – Differentially Private Autoregressive Tabular, a General Framework for Synthetic Data Generation (published at TPDP 2022)
• py-synthpop – Python implementation of the R package synthpop

I maintain a public repo containing a collection of Differentially Private (tabular) Generative Models Papers with Code. Please contribute!

I also regularly contribute to open source projects, mainly related to privacy-preserving machine learning.