Astronomer

Overview

Astronomer is not a separate orchestration engine. It is the main commercial vendor of managed Apache Airflow, and its product, Astro, runs Airflow for you. Existing DAGs run unchanged, which is the whole proposition: keep Airflow, stop operating it.

What Astronomer adds is the surrounding machinery. Astro Runtime is its packaged Airflow image; the hypervisor scales deployments and clusters; a browser-based IDE and AI assistance help write DAGs; and the CLI handles local development and deploys.

The part worth attention when comparing is Astro Observe. It lets you define data products, compositions of pipelines and datasets that span deployments, then shows real-time lineage and raises alerts when a freshness or delivery target is missed. OpenLineage is configured automatically, which is a real saving over wiring it up yourself.

Execution can be hosted, where Astronomer manages workers, autoscaling and provisioning, or remote, where tasks run in your own infrastructure and only orchestration stays on Astro. Remote execution is an enterprise-tier feature. Worker sizes run from one to 32 vCPUs.

Astronomer positions itself as Airflow's commercial steward and says it employs 18 of the project's top committers and 10 PMC members, which is worth knowing both as a sign of investment and as a concentration of influence in one vendor, though it holds no governance authority over the Apache project.

It runs on AWS, Google Cloud and Azure, with an Astro Private Cloud option for self-managed installations.

Key features and capabilities

The same headings are used for every orchestration & scheduling entry, so two tools can be read side by side.

How pipelines are written

• Standard Airflow Python DAGs, unchanged from open source
• Astro IDE, a browser-based development environment, with AI-assisted authoring
• Astro CLI for local development and deploys
• Documented CI/CD automation for code deploys

Scheduling and triggers

• Inherits Airflow 3 scheduling, including assets, event triggers and backfills
• Service-level targets on data products raise alerts when freshness slips
• Deployments can hibernate and scale compute to zero on the entry tier

Execution and scaling

• Hosted execution, where Astronomer manages workers, autoscaling and provisioning
• Remote execution runs tasks in your own infrastructure, on enterprise tiers
• Standard multi-tenant clusters or dedicated single-tenant clusters
• Seven worker sizes, from 1 vCPU and 2 GiB up to 32 vCPU and 64 GiB

Monitoring and recovery

• Astro Observe defines data products spanning pipelines and datasets across deployments
• Real-time lineage graph of assets and their upstream dependencies
• Alerts and targets consolidated per data product, with notification history
• Audit logging of 7 or 90 days depending on tier

Data awareness

• Airflow assets and asset-based scheduling carry over
• OpenLineage is configured automatically to feed observability and alerts
• Data products infer their upstream dependencies
• Data quality features are part of the Team tier

Integrations

• The full Airflow provider ecosystem, over 100 packages
• dbt Cloud, Spark, Snowflake and the major clouds among them
• An MCP server for its documentation

How it runs

• Managed on AWS, Google Cloud and Azure
• Standard clusters in a handful of regions; dedicated clusters in many more
• Astro Private Cloud for self-managed installation
• Proprietary, running Apache 2.0 Airflow underneath

Pricing

Usage-basedDeployments from $0.35 an hour

Developer starts at $0.35 an hour per deployment, pay as you go, with AI-assisted authoring and scale-to-zero compute. Team starts at $0.42 an hour and adds observability, data quality, dedicated clusters and 24x5 support. Business, Enterprise and Business Critical are quoted annually, adding single sign-on enforcement, remote execution, custom roles and faster support commitments. Workers cost from $0.13 an hour and dedicated clusters from $2.40 an hour. A free trial is offered but its length is not published.

Vendor pricing page →

Demos and videos

• How to get started with Astro Tutorial · YouTube
• Astro Observe: orchestration-native observability for Airflow Product demo · YouTube

About Astronomer

Astronomer sells managed Apache Airflow and moved its head office from Cincinnati to New York in 2024. It raised $93m in a Series D in May 2025 led by Bain Capital Ventures, with Salesforce Ventures, Insight, Meritech and others, following a $213m Series C that accompanied its acquisition of Datakin, the company behind OpenLineage. It does not publish a founding year, saying only that it has driven Airflow development since 2018, and cites more than 700 enterprise customers. Astro is proprietary; the Airflow underneath is Apache 2.0.

New York, New York · astronomer.io