Agent engine quick starts

These tutorials get you started using Airbyte's Agent Engine and agent connectors. They're intended for people new to Airbyte's Agent Engine and its agent connectors. Follow these tutorials to go from having nothing to having a locally running AI agent that can work with your data in real time. In most cases, you can achieve this in under half an hour.

📄️ Pydantic AI

In this tutorial, you'll create a new Python project with uv, add a Pydantic AI agent, equip it to use one of Airbyte's agent connectors, and use natural language to explore your data. This tutorial uses GitHub, but if you don't have a GitHub account, you can use one of Airbyte's other agent connectors and perform different operations.

📄️ LangChain

In this tutorial, you'll create a new Python project with uv, add a LangChain agent, equip it to use one of Airbyte's agent connectors, and use natural language to explore your data. This tutorial uses GitHub, but if you don't have a GitHub account, you can use one of Airbyte's other agent connectors and perform different operations.

📄️ FastMCP

In this tutorial, you'll create a new Python project with uv, build a FastMCP server that exposes one of Airbyte's agent connectors as an MCP tool, and use it to query GitHub data from any MCP-compatible agent. This tutorial uses GitHub, but if you don't have a GitHub account, you can use one of Airbyte's other agent connectors and perform different operations.

📄️ Agent Engine MCP

In this tutorial, you install Airbyte's Agent Engine MCP server, configure a connector, and register the server with your agent. Once registered, you can use natural language to query your data sources. You don't have to write any code.

📄️ Hosted connectors

When you run connector operations with the Python SDK, you store API credentials locally and provide them directly to the API through agent connectors. This approach, while viable at first, reaches limits quickly. You may find yourself dealing with large numbers of customers who have their own environments and credentials. You may not want to manage these credentials and or store them at all.