dbt
The technology
dbt (data build tool) lets data transformations be written in versioned, tested and documented SQL, directly in the warehouse: Snowflake, Databricks, BigQuery or Fabric.
dbt brings software engineering practices to data transformation. They still have to be kept alive: modelling conventions, systematic tests, code review and up-to-date documentation.
Engagements
Visian engagements on dbt
- Setting up a dbt project and modelling conventions
- Taking over existing transformations (stored procedures, ETL)
- Quality tests, data contracts and generated documentation
- Orchestration (Airflow, dbt Cloud, Dagster) and CI/CD
- Training data engineering and analytics engineering teams
These engagements are part of the Data platformexpertise, on a fixed-price, time-and-materials or service-centre basis.
Frequently asked questions
dbt: FAQ
Does dbt replace an ETL tool?
dbt covers transformation (the “T”). Extraction and loading are still handled by another tool: Fivetran, Airbyte, Azure Data Factory or scripts.
dbt Core or dbt Cloud?
dbt Core is open source and fits into existing orchestration. dbt Cloud adds the IDE, scheduling and hosted documentation, at a per-user cost.