Databricks Asset Bundles Example with uv

This project shows an example of a Python project that uses uv for dependency management and Databricks Asset Bundles for deployment.

Local development

Prerequisites:

• Python 3.9+
• uv
• Databricks CLI
• Databricks workspace with Unity Catalog and Serverless enabled

Steps:

1. Clone the repository
2. Setup the uv env:

uv sync

Running tests from local machine

Add the DATABRICKS_CONFIG_PROFILE environment variable to your .env file:

DATABRICKS_CONFIG_PROFILE=<your-profile-name>

Then, run the tests via pytest:

pytest

Deployment

To deploy the app, login to your Databricks workspace from Databricks CLI:

databricks auth login 

Deploy the workflow:

databricks bundle deploy --var="catalog=your_catalog" --var="schema=default" -t dev

Tech used

• Databricks Asset Bundles - The tool to create, manage, and deploy Databricks Workflows
• uv - An extremely fast Python package and project manager, written in Rust
• hatch-vcs - Plugin that uses your preferred version control system (like Git) to determine project versions
• pydantic - data validation library for Python