dbt-redshift

dbt-redshift is the official dbt adapter that connects dbt Core to Amazon Redshift. It provides the connection layer, Redshift SQL dialect, and materialization logic dbt needs to compile and execute models, tests, snapshots, and seeds against a Redshift cluster or Redshift Serverless.

Maintained by dbt Labs within the dbt-adapters monorepo, it builds on dbt-postgres for shared Postgres-compatible behavior while adding Redshift-specific connection handling via the redshift-connector. Installing it alongside dbt Core lets an analytics-engineering project target Amazon Redshift.

What You Get

• A dbt adapter that targets Amazon Redshift as the execution engine for dbt models
• Redshift connection and authentication handling, including IAM auth providers
• Redshift-aware materializations built on top of dbt-postgres behavior
• Relation configuration mapped to Redshift-specific features
• Integration with the standard dbt Core CLI workflow (run, test, build, snapshot)

Common Use Cases

• Running a dbt analytics-engineering project against an Amazon Redshift cluster
• Building incremental and materialized transformations in Redshift with dbt
• Managing Redshift credentials and IAM-based auth through dbt profiles

Under The Hood

Architecture - The adapter lives under dbt-redshift/src/dbt/adapters/redshift/. connections.py manages the Redshift connection and query execution, impl.py implements the RedshiftAdapter dbt Core calls into, auth_providers.py handles IAM-based authentication, and relation.py plus relation_configs/ describe how dbt relations map to Redshift objects. Because Redshift is Postgres-derived, the package depends on and reuses much of dbt-postgres.

Tech Stack - Python packaged with Hatch and pyproject.toml. It builds on dbt-common, dbt-adapters, and dbt-postgres for the plugin contract and shared behavior, and depends on redshift-connector for the wire connection plus sqlparse and dbt-core for installation compatibility.

Code Quality - The package includes unit and functional tests under tests/, uses black and mypy, and keeps a detailed CHANGELOG. As part of the actively maintained dbt-adapters monorepo it shares CI and the reusable dbt-tests-adapter suite. The pinning notes in pyproject reflect careful handling of redshift-connector’s non-SemVer releases.

API Design - Users interact through dbt’s YAML profiles and SQL models rather than the Python API. Pointing dbt at Redshift is a profile-block change, and the adapter follows the uniform dbt adapter contract, so experience transfers directly from other warehouses.