Skip to content

How It Works

These pages are the fast narrative path to a working mental model of tatara: the shape of the system and why it is built that way, for architects and operators sizing up whether to adopt or run it. Start here for the big picture and the load-bearing design choices, then drop into the reference section when you need field-level detail.

  • The Big Picture

    What tatara is, why it exists, and how all the pieces fit together in one narrative.

    The Big Picture

  • From Issue to PR

    Walk through exactly what happens when you open a GitHub issue with the tatara label.

    From Issue to PR

  • The Memory Graph

    How tatara builds and queries a knowledge graph of your codebase.

    The Memory Graph

  • Agent Sessions Explained

    What is a tatara agent session? How does Claude Code run inside a Kubernetes pod?

    Agent Sessions