Local container adds Kafka support

Tinybird Local, our local container, now supports Kafka data sources. You can follow same process as you follow with regular data sources. See the docs for an example.

By the way, the new default port of the container is 7181. We're still exposing port 80 for backward compatibility, but we'll remove it in 2 weeks. Change ports now!

Tinybird Forward - Weekly update

Here's a selection of the weekly updates we post in the #forward channel:

  • You can now use forward queries in materialized data sources to avoid losing old materialized data through recalculations. See the docs for an example.
  • You can now place your .tinyb file anywhere in your project. Tinybird CLI looks for the .tinyb file in all parent directories till it reaches your home.
  • Projects that use organization service data sources are now supported.
  • Lots of small improvements in the Tinybird Cloud UI – it’s getting better every day!

Read the blog post and browse the docs to learn more.

Better onboarding experience

When creating a new workspace, the onboarding screen guides you through the process of creating your first data source, query data with a pipe, and publish the data with an API endpoint.

When you're done, the onboarding now leads you to your dashboard, where you can immediately check activity in your endpoints and get other valuable insights.

Dashboard

Email notifications for playground sharing

As you know, you can share your playgrounds with other users. Now, when you share a playground, the user you shared it with will receive an email notification. The same happens when the playground stops being shared.

Improvements

  • Orgs on the new pricing plans can now check vCPU usage and active minutes consumed using the organization.shared_infra_active_minutes service data source.
  • You can now pick which organization to use when creating a new workspace from the CLI.
  • Standardized error messages in datasources_ops_log data source.
  • Added CLI support for Python 3.13.

Bug fixes

  • We added checks to prevent the creation of a materialized view with the same target and origin data source, thus avoiding recursion errors during ingestion.
  • We fixed some issues that were causing users to get stuck in the 404 page when trying to access the UI.