Give your data team magical powers.

For documentation on getting started, how to develop, and how to deploy to production check out the live
Developer documentation portal.

The recommended way to install the latest version of Mage is through Docker with the following command:

docker pull mageai/mageai:latest

You can also install Mage using pip or conda, though this may cause dependency issues without the proper environment.

pip install mage-ai

conda install -c conda-forge mage-ai

Looking for help? The fastest way to get started is by checking out our documentation here.

Looking for quick examples? Open a demo project right in your browser or check out our guides.

Build and run a data pipeline with our demo app.

WARNING

The live demo is public to everyone, please don’t save anything sensitive (e.g. passwords, secrets, etc).

_{Click the image to play video}

A sample data pipeline defined across 3 files ➝

1. Load data ➝

   @data_loader
   def load_csv_from_file() -> pl.DataFrame:
       return pl.read_csv('default_repo/titanic.csv')

2. Transform data ➝

   @transformer
   def select_columns_from_df(df: pl.DataFrame, *args) -> pl.DataFrame:
       return df[['Age', 'Fare', 'Survived']]

3. Export data ➝

   @data_exporter
   def export_titanic_data_to_disk(df: pl.DataFrame) -> None:
       df.to_csv('default_repo/titanic_transformed.csv')