Work with data

When your data is in Tinybird, you can process it and explore it in different ways. The main way to do it is via pipes.

  • Pipes are a way to query and transform data. You can read data from a data source, apply transformations, and use the result from another pipe, like an endpoint. Pipes can be specialized into endpoints, materialized views, copy pipes, and so on.

Publish data

When your data is in Tinybird, you can publish it as API Endpoints.

  • Endpoints are a way to publish data as REST API Endpoints. They can be used in your frontend, backend, or any other application and can be secured with JWT.

Optimize

Tinybird provides several ways to process and copy data between data sources to optimize your endpoints:

  • Copy pipes capture the result of a pipe at a specific point in time and write it to a target data source. They can run on a schedule or run on demand, making them ideal for event-sourced snapshots, data experimentation, and deduplication with snapshots.

  • Materialized views continuously re-evaluate a query as new events are inserted, maintaining an always up-to-date derived dataset. Unlike copy pipes which create point-in-time snapshots, materialized views provide real-time transformations of your data.

Each approach has its own strengths and use cases:

  • Use copy pipes when you need scheduled or on-demand snapshots of your data.
  • Use materialized views when you need continuous, real-time transformations.

Lineage

Lineage visualizes how your data sources, endpoints, materialized views, and pipes connect and relate to each other.

You can select an item to see its details in the side panel.

Explorations

Explorations is a conversational UI feature in Tinybird that lets you explore and interact with your data using natural language.

MCP server

The Tinybird MCP server enables AI agents to connect directly to your Tinybird workspace to execute queries or use your endpoints as tools using the Model Context Protocol.

Next steps

Updated