Data engineering teams building and maintaining data pipelines and ML feature pipelines.

Dagster is a cloud-native data orchestration platform with a declarative programming model, built-in lineage and observability for data assets.

Self-hostable open source core, with Dagster Plus available as a managed cloud offering.

Actively maintained with nearly 16,000 GitHub stars and close to 28,000 commits into 2026.

Treats data pipelines as testable software assets with lineage tracking, making it easier to debug and trust pipeline output.

Written in Python and integrates with dbt, Spark, Pandas and most common data warehouses and cloud storage.

Apache 2.0 open source and self-hostable, Dagster Plus is an optional paid managed cloud service run by Dagster Labs.