Introduction

PyAirbyte wraps the entire Airbyte connector catalogue in a friendly Python API. Instead of running the Airbyte platform you can pull data right inside a Jupyter/Colab notebook, a scheduled script, or a production service. In this course we will:

• Install and configure PyAirbyte.
• Build a pipeline that pulls data from Stripe into DuckDB.
• Persist data in MotherDuck and keep it incrementally updated.
• Run a custom manifest‑only connector built with the Airbyte Connector Builder.
• Load data into a vector database and run semantic queries (RAG).

Along the way we will cover secret management, caching options, and deployment tips.

 Prerequisites

[!Tip] All examples run in Colab so you can follow even on an iPad.