Skip to main content

Build modern data pipelines with confidence

Mage is an open-source data pipeline tool for building, running, and orchestrating data transformations. Create ETL/ELT workflows with Python, SQL, or R in a visual notebook interface.

Get StartedView API Reference

Quick start

Get up and running with Mage in minutes

1

Install Mage

Install Mage using pip, Docker, or conda:
pip install mage-ai
2

Initialize your project

Create a new Mage project in your desired directory:
mage init my_first_project
cd my_first_project
This creates a new project with a default folder structure including pipelines, data loaders, transformers, and exporters.
3

Start the Mage server

Launch the Mage UI and server:
mage start
Open your browser and navigate to http://localhost:6789 to access the Mage interface.
By default, Mage runs on port 6789. You can customize the host and port using mage start --host 0.0.0.0 --port 8080
4

Create your first pipeline

In the Mage UI:
  1. Click + New pipeline
  2. Choose a pipeline type (Standard, Streaming, or Integration)
  3. Add blocks to load, transform, and export your data
  4. Run your pipeline and view the results
See the Creating Pipelines guide for detailed instructions.

Explore by topic

Learn about Mage’s key features and capabilities

Pipelines

Build data pipelines with modular blocks using Python, SQL, or R

Data Integrations

Connect to 50+ data sources and destinations out of the box

Orchestration

Schedule and monitor pipeline runs with triggers and sensors

Development

Write and test code in an interactive notebook-style interface

Deployment

Deploy to Docker, Kubernetes, AWS, GCP, Azure, and more

Advanced Features

Explore dbt integration, streaming, Spark, and AI features

Popular integrations

Connect to your data sources and destinations

Database Sources

PostgreSQL, MySQL, BigQuery, Snowflake, Redshift, and more

View sources

API Sources

REST APIs, GraphQL, Salesforce, Stripe, Google Analytics, and more

View sources

Cloud Storage

S3, GCS, Azure Blob Storage, and local file systems

View sources

Streaming Sources

Kafka, RabbitMQ, AWS Kinesis, Google Pub/Sub, and more

View sources

Key features

What makes Mage powerful for data engineering teams

Notebook-style interface

Write and test code in an interactive notebook environment with instant feedback and data previews

Dynamic blocks

Execute blocks in parallel based on runtime conditions for efficient data processing

Built-in orchestration

Schedule pipelines with cron expressions, event triggers, or API calls

AI-powered development

Generate code, fix errors, and optimize queries with AI assistance

Ready to build your first pipeline?

Get started with Mage in minutes and transform your data workflows with modern orchestration tools.

Start building now

Build docs developers (and LLMs) love

Get started for freeTalk to us