dbt

jpoutrin's avatarfrom jpoutrin

dbt (data build tool) patterns for data transformation and analytics engineering. Use when building data models, implementing data quality tests, or managing data transformation pipelines.

⭐0stars🔀0forks📁View on GitHub🕐Updated Dec 9, 2025
[SQL and Databases]

When & Why to Use This Skill

This Claude skill provides comprehensive dbt (data build tool) patterns and best practices for analytics engineering. It enables users to architect robust data transformation pipelines, implement automated data quality testing, and manage complex SQL-based data models using industry-standard staging, intermediate, and marts structures to ensure data reliability and maintainability.

Use Cases

  • Architecting Data Pipelines: Designing multi-layered data models to transform raw warehouse data into clean, business-ready insights using standardized staging and marts patterns.
  • Automated Data Quality Testing: Implementing schema tests and data assertions (like unique and not_null) to ensure the integrity and reliability of analytics datasets.
  • Incremental Data Loading: Developing efficient incremental models to process large-scale datasets by only updating new or changed records, reducing warehouse compute costs.
  • Standardizing Analytics Workflows: Applying modular SQL development practices, including source referencing and macro usage, to improve code reusability and team collaboration.
namedbt
descriptiondbt (data build tool) patterns for data transformation and analytics engineering. Use when building data models, implementing data quality tests, or managing data transformation pipelines.

dbt Skill

This skill provides dbt patterns for analytics engineering.

Project Structure

dbt_project/
├── dbt_project.yml
├── models/
│   ├── staging/
│   │   └── stg_customers.sql
│   ├── intermediate/
│   │   └── int_customer_orders.sql
│   └── marts/
│       └── fct_orders.sql
├── seeds/
├── macros/
├── tests/
└── snapshots/

Model Patterns

Staging Models

-- models/staging/stg_customers.sql
with source as (
    select * from {{ source('raw', 'customers') }}
),

renamed as (
    select
        id as customer_id,
        lower(email) as email,
        created_at
    from source
)

select * from renamed

Incremental Models

-- models/marts/fct_orders.sql
{{
    config(
        materialized='incremental',
        unique_key='order_id'
    )
}}

select *
from {{ ref('stg_orders') }}
{% if is_incremental() %}
where updated_at > (select max(updated_at) from {{ this }})
{% endif %}

Testing

# models/schema.yml
models:
  - name: stg_customers
    columns:
      - name: customer_id
        tests:
          - unique
          - not_null
      - name: email
        tests:
          - unique

Best Practices

  • Use staging → intermediate → marts pattern
  • Source all raw data with source()
  • Reference models with ref()
  • Add documentation and tests
  • Use incremental models for large datasets