What is dbt™?

dbt™ (data build tool) is the open-source framework Paradime is built around. If you’re new to it, this page gives you just enough to understand the rest of the docs.

​

The core idea

dbt™ lets you transform data inside your warehouse using SQL and version-controlled code. Instead of fragile scripts or point-and-click pipelines, your transformations live in a git repository as a dbt™ project: reviewable, testable, and repeatable. You write SQL SELECT statements; dbt™ handles turning them into tables and views, running them in the right order, and testing the results.

​

The building blocks

• Sources: declarations of the raw tables your warehouse already contains, so your project knows where data comes from.
• Models: SQL files that transform data. Each model becomes a table or view in your warehouse. Models reference each other, and dbt™ works out the execution order automatically.
• Tests: assertions about your data, like “this column is never null” or “these values are unique”. They catch quality issues before your stakeholders do.
• Documentation: descriptions of models and columns that live next to the code and power the Catalog.

​

A typical flow

1. Declare your raw tables as sources.
2. Build staging models that clean them up.
3. Build marts models that answer business questions.
4. Add tests and documentation along the way.
5. Run the whole thing on a schedule in production.

In Paradime, steps 1 to 4 happen in the Code IDE (with DinoAI accelerating the work), and step 5 is Bolt’s job.

​

Where to go next