Course Wiki

dbt

SQL-first analytics engineering: transformations as versioned, tested, documented models with a dependency graph.

Module 4 is the analytics engineering module, and dbt is its tool: transformations from raw warehouse data to analytical models are SELECT statements in files, organized into a project with a dependency graph (refs between models), tests (uniqueness, not-null, accepted values), and generated documentation.

The module builds a dbt project over the NYC taxi data on two stacks — locally with DuckDB and dbt Core, or in the cloud with BigQuery and dbt Cloud — so learners see that the project structure is portable across warehouses. The outcome discipline: transformations that are version-controlled, tested, and documented instead of buried in a notebook.

Taught in

• Data Engineering Zoomcamp

  • Module 4: Analytics Engineering
    • Analytics Engineering ## Related concepts

• BigQuery

• Partitioning and Clustering

• Analytics Engineering

• Bruin

• Workflow Orchestration