Concepts overview

Before diving into features, it helps to understand the handful of objects that everything in Zingle AI is built on. They map closely to how dbt teams already think about their work.

Concept	What it is
Organization	The tenancy root. Every user, repo, and connection belongs to exactly one org.
User & role	People in an org. Roles gate what they can do (admin vs analyst).
Repository	A registered dbt GitHub repo Zingle AI can read and open PRs against.
Environment	A named binding of a repository to a warehouse compute target.
Connection & target	A warehouse (Snowflake) connection and the compute targets within it.
dbt model & semantic layer	The synced catalog of models, columns, lineage, and semantic definitions.
Agent	An LLM-driven worker (Build, SQL, Refactor, Optimize, or custom) with a prompt, tools, and skills.

Each concept has its own page in this section. Start with Organizations & roles.