11% domain · module 1
Core platform: lakehouse, Delta, Lakeflow, Mosaic AI, DBSQL
Starts at 28:17 in this video — SQL Warehouse full course — Databricks overview (28:17)
What the Oct 2025 exam actually lists
The official outline asks you to describe the core components of the Databricks Data Intelligence Platform: Mosaic AI, Lakeflow (Declarative Pipelines — formerly Delta Live Tables — and Lakeflow Jobs), the Data Intelligence Engine, Delta Lake, Unity Catalog, and Databricks SQL. You are not building Spark jobs on the exam, but you must know which layer owns which job.
How the pieces fit for an analyst
- Delta Lake — ACID tables on cloud object storage (Parquet + transaction log). Time travel,
MERGE, schema enforcement live here. - Unity Catalog — one governance plane: identity, privileges, lineage, comments, volumes. Three-level names:
catalog.schema.table. - Databricks SQL — the analyst workspace: SQL editor, warehouses, AI/BI dashboards, alerts, Genie.
- Lakeflow Declarative Pipelines (DLT) — engineers declare bronze/silver/gold pipelines. Analysts consume the gold tables.
- Lakeflow Jobs — schedules notebooks, SQL, pipelines, dashboards. (Older exam blogs still say “Workflows”.)
- Mosaic AI — model serving, agents, vector search. Analysts mostly *use* AI/BI (Genie, Assistant), not train models.
- Data Intelligence Engine — the platform’s AI that uses UC metadata, query history, and lineage so Assistant and Genie know *your* tables.
Compute vs storage
Storage is cheap object storage. Compute is attached on demand. Analysts attach a SQL warehouse, not an all-purpose cluster, for SELECT / dashboards / Genie.
Exam trap
If a question mixes “where is the data stored?” vs “where do I run SQL?” vs “who is allowed to see it?”, answer Delta/cloud storage, SQL warehouse, and Unity Catalog respectively.