What is synthetic data?

In its most basic form, synthetic data is simply a collection of samples generated using a program. As these generated samples do not represent living entities, synthetic data is free from privacy risks. Our technology specialises in powerful generative models that synthesize tabular data which closely mimic the properties of original data to unlock sensitive data to be shared quickly and reliably without breaching the privacy of individuals.

What do we offer?

Strong Privacy Enforcement 

Generatrix provides 2-layers of privacy protection with our novel privacy-by-design training framework and formal mathematical privacy guarantees ensuring an extremely secure synthetic data generation pipeline where the real data never leaves the user's work environment.

Dependable Performance 

Generatrix remains unfazed in the presence of missing values and is specially designed to reliably reproduce biased and heavily skewed data columns. Our system efficiently processes high-dimensional categorical columns with 100s of categories effortlessly.

Flexible Integration 

Generatrix delivers familiar usability and easily integrates into your organisation without disrupting your workflow. It can be broken down into the following two main components:

Desktop App

  • Easy install for macOS, Windows and Linux.

  • Intuitive UI design with a minimal learning curve

Library Toolkit 

  • Easily merges with existing ML infrastructure 

  • Ready to deploy on a range of cloud services​

Comprehensive Reporting

Generatrix constantly assesses the generation quality and showcases a report visually comparing the real and synthetic data. Our report highlights a detailed analysis of the utility for training machine learning algorithms and extensively measures the risk of privacy leaks.