Dataiku was introduced to Maersk as a replacement for Palantir Foundry. DSS covers powerful capabilities in terms of data modelling and wrangling. In addition, it lowers the bar of entry to Machine Learning for everyone.
Create and collaborate on a project, connect your external tools, and enjoy exploring, modelling and wrangling data, and enhance with machine learning capabilities, just as you like it.
Sort, model, and analyze your chosen data from Maestro and utilize the versatile offering of tools inside of DSS that allows for machine learning capabilities.
Explore the Maestro data lake, re-cycle existing data models and save time on your data discovery and development work
Collaborate across business units and tap into the community of knowledge sharing - exactly like your know from i.e. Github & Palantir Foundry
Below you can find information about who is the typical user of Dataiku DSS and how to get started.
Analysts, Engineers & Data Scientists, who aspire to do anything related to data in a consolidated space, can utilize and unleash the power of Dataiku DSS. As DSS contains the tools needed to explore and utilize the Maestro data pool in an easy and seamless manner, you will be able to get new insights, build and wrangle your data model to test out your hypotheses in a quick and collaborative style. DSS also offers you endless possibilities when it comes to connecting your external tools of choice.
Databricks is provided through either a shared workspace or a dedicated workspace. If you need a dedicated workspace, you need to register a product in Maestro and get the workspace in the registered resource group. Further, for devops pipelines, you could use Maestro Application to set up your Project and Databricks Repository in Azure devops as well as code deployment Pipelines.