1. Introduction
You'll map the platform layers you need to build, in the order they depend on each other, in ~10 min.
Prereqs: A Databricks account
Start with the foundation. Every dashboard, model, and agent reads data from the same platform and catalog. If the tables are wrong, reports mislead, models train on noise, and agents return unreliable answers.
BI, machine learning, and AI have different serving needs. They still depend on the same account, governance model, and trusted data. Fix those layers once, then build each workload on top.
This guide follows that dependency order. It starts with the account and workspaces, moves through data access and governance, then reaches the products people use. Operations and CI/CD closes the loop by making the platform repeatable.
Starter Journey Progress
What you'll build
| # | Section | What you'll have when done |
|---|---|---|
| 1 | Introduction | A foundation-first map of the journey and a cloud tenant model that fits your organization |
| 2 | Account and workspaces | Workspaces, identity, SSO, account administration, workspace settings, and cost monitoring |
| 3 | Data access and ETL | Connected data sources, a working ETL pipeline, SQL exploration, and scheduled jobs |
| 4 | Governance | A Unity Catalog structure, ownership model, and access controls for your data |
| 5 | Genie Ontology | Generated metadata, governed metrics, dashboards, Genie agents, and a domain structure |
| 6 | Data science | Governed feature tables, registered models, and batch inference |
| 7 | Operations and CI/CD | Version-controlled Databricks projects deployed with DABs and CI/CD |
Next
- Do next: Foundations
- Learn why: Cloud tenant ready
- Reference: Databricks architecture overview