Repolode
Submit
dagster-io

dagster-io/dagster

A Python orchestration platform for defining, running, and monitoring data assets such as tables, datasets, ML models, and reports.

Open-source alternative to:Apache Airflow

View on GitHubWebsite
Stars
16,311
Forks
2,366
Open issues
2,604
Last commit
10 hours ago

What is it?

Dagster lets you declare the data assets you want to build as Python functions, then handles running them at the right time and keeping them up to date. Assets are connected into a graph, so dependencies, lineage, and run history are visible in its web UI.

The project is designed to be used across the whole development lifecycle, from local development and unit tests to integration tests, staging, and production. It provides scheduling, multi-tenant execution, built-in observability and cataloging, and a growing library of integrations with common data tools.

It supports Python 3.9 through 3.14 and is installable from PyPI with the dagster and dagster-webserver packages.

Key features

  • 01Define data assets as Python functions with a declarative model.
  • 02Visualize asset dependencies and lineage in a web UI.
  • 03Schedule and run assets, keeping them up to date.
  • 04Built-in observability, diagnostics, and cataloging of metadata.
  • 05Multi-tenant orchestration engine for production deployments.
  • 06Growing library of integrations with popular data tools.

Use cases

  • Building ETL or ELT pipelines that produce warehouse tables.
  • Orchestrating machine learning training and model pipelines.
  • Maintaining reports and datasets that must stay up to date.
  • Testing data pipelines locally and in CI before production rollout.

Quick start

  1. 1

    Install Dagster, the web server and the CLI with uv

    uv add dagster dagster-webserver dagster-dg-cli
  2. 2

    Follow the hands-on tutorial in the docs to build your first pipeline

Topics

#data-engineering#orchestration#etl#workflow#data-pipelines#python#mlops#metadata#analytics#dagster

FAQ

What is dagster?
A Python orchestration platform for defining, running, and monitoring data assets such as tables, datasets, ML models, and reports. Dagster lets you declare the data assets you want to build as Python functions, then handles running them at the right time and keeping them up to date. Assets are connected into a graph, so dependencies, lineage, and run history are visible in its web UI.
Is dagster free to use? What license does it have?
dagster is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is dagster still maintained?
Actively maintained. The last commit to dagster was on Oct 21, 2026.
What can I use dagster for?
Building ETL or ELT pipelines that produce warehouse tables. Orchestrating machine learning training and model pipelines. Maintaining reports and datasets that must stay up to date. Testing data pipelines locally and in CI before production rollout.
How do I install dagster?
Install Dagster, the web server and the CLI with uv: uv add dagster dagster-webserver dagster-dg-cli Follow the hands-on tutorial in the docs to build your first pipeline
Where can I get dagster?
The source code and installation instructions are on GitHub: https://github.com/dagster-io/dagster

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects