Dataiku APIs

• Using the internal Python API, a Data Scientist can read a Dataiku DSS dataset into a dataframe, process the dataframe with Python code, and then write back to a Dataiku DSS dataset. Examples include:
  • Reading a Dataiku DSS dataset to create a Bokeh web app,
  • How to execute an SQL query from a Python recipe to read a Dataiku DSS dataset into a dataframe

Using the internal R API, a Data Scientist can read a Dataiku DSS dataset into a dataframe, process the dataframe with R code, and then write back to a Dataiku DSS dataset. Examples include:

• Reading a Dataiku dataset to create a Shiny web app
• Reading a Dataiku dataset to create an RMarkdown report

Using the internal Javascript API, a Data Scientist can read a Dataiku DSS dataset into a dataframe. Examples include:

• Reading a Dataiku dataset to create a Javascript/CSS web app

A Data Scientist can create custom scenarios using the internal Python API.

An Application Developer can query scoring services on the Dataiku API node.

Using the Public API, a Production Environment Manager can:

• Download project bundles from a Design node,
• Upload them to and manage them on an Automation node, and
• Transfer API service packages from an Automation node to an API node.

A Data Scientist can create plugins that contain custom recipes, datasets, and processors, using mostly the Python API, though the R API can be used for custom recipes. Examples include:

• Writing your first Dataiku DSS plugin

• Manage security settings, such as creating users, groups, and projects, on a Dataiku DSS instance
• Manage connections from Dataiku to various data stores
• Populate a project with datasets, recipes, and models according to preset configurations

In this way, you can spin up and take down Dataiku instances as they are needed on cloud infrastructures.