Djinn is a version of Tonic that is intended specifically for data scientists.
In Djinn, you create models. For each model, you configure a SQL query to retrieve a subset of data for Djinn to transform and to train. The training process uses the Tonic AI Synthesizer to generate synthetic data.
A model provides a view of the underlying data that is specified by the SQL query. It represents a downstream analysis or a data science task to answer a specific question.
You export the results to a Jupyter notebook, and use Jupyter analysis and visualizations to verify that the transformed data corresponds accurately to the source data. You can also use the trained data for other analysis.
At a high level, in Djinn, you: