dbt Data Transformations
Author dbt models, tests, and docs for analytics warehouses. Generates staging/mart layers, incremental models, snapshot logic, ref/source patterns, and CI tests for Snowflake, BigQuery, and Databricks.
dbt is the standard for ELT transformation in the warehouse. This skill scaffolds project structures with staging/intermediate/marts layers, writes incremental and snapshot models, adds generic and singular tests, and wires dbt-checkpoint into CI for PR enforcement.
dbt data-modeling elt sql analytics-engineering
When to use
Use when SQL is fragmenting across notebooks and stored procs — introduce a versioned, tested, documented transformation layer your analytics team can own.
Examples
Incremental model with late-arriving data
Handle backfills without full refreshes
Write an incremental dbt model for orders that handles late-arriving events using a lookback window and merge strategy
Staging + marts layering
Restructure a flat models folder
Our dbt project has 80 models in one folder — restructure into staging/intermediate/marts with proper ref usage and yml docs