Skip to main content

10. Databricks AI/BI

You'll learn how Databricks AI/BI turns lakehouse data into dashboards, natural-language analytics, and custom applications in ~5 min.

Prereqs: Query and explore

Why this matters​

Data sitting in tables is useful to engineers. Business users need dashboards, self-service Q&A, and purpose-built applications. Databricks AI/BI is the presentation layer of the lakehouse — it connects directly to Unity Catalog tables, respects row- and column-level security, and removes the need to export data to a separate BI tool.

Journey checklist​

  • Get started.
  • Before you start.
  • Infra setup.
  • Cost monitoring.
  • Data Governance Strategy.
  • Access your data.
  • Build the first pipeline.
  • Automation and orchestration.
  • Query and explore.
  • Databricks AI/BI.
  • Business semantics.

How it works​

Databricks AI/BI has three components, each targeting a different audience:

ComponentAudiencePurpose
DashboardsAnalysts, stakeholdersVisual reports with charts, filters, and scheduled refreshes
Genie SpacesBusiness usersNatural-language interface — ask questions about data in plain English
Databricks AppsDevelopersCustom web applications hosted on Databricks with built-in auth and data access

All three read from the same governed tables. A dashboard, a Genie Space, and an app can point at the same gold-layer table — one source of truth, three consumption patterns.

When to use / when not to​

Use Databricks AI/BI when:

  • You want a single platform for data engineering and data presentation.
  • Governance (row/column security, lineage) must extend to the BI layer.
  • Business users need self-service analytics without learning SQL.

Use an external BI tool when:

  • Your organization has standardized on a third-party tool (Tableau, Power BI) and migration is not planned.
  • You need features specific to that tool (e.g., Tableau's spatial visualizations).

In this section​

  • Dashboards — Create and share visual reports.
  • Genie Spaces — Let business users ask data questions in natural language.
  • Databricks Apps — Build and deploy custom web applications.

Next​