dbt Wizard: An agent purpose-built for analytics engineering

🌎 Global friendly sessions on July 22 & 23

Most AI coding agents are generalists. They know Python. They know SQL. They don't know your DAG, your model dependencies, your failing job, or why your incremental model broke in production at 3am. Even with MCP, generic agents are stitching together context they were never built to understand.

dbt Wizard is built for the analytics engineer. It works alongside you with deep domain knowledge and a complete understanding of your project and workflows. It knows which model is right, what a change will break, which tool to call, and when to validate — with governance and guardrails on by default.

In this deep-dive, the team demos dbt Wizard across two surfaces — the CLI for engineers who live in the terminal, and the dbt platform for teams that want a shared interface for exploratory questions, failed job remediation, and visual DAG navigation. We'll show you real use cases — building new models, debugging failures, refactoring with dbt compare, and authoring tests — and walk through how analytics teams are using it today.

Meet the speakers

Arjit Jaiswal

Senior Product Manager

Fivetran

Brandon Thomson

Senior Manager, Data Engineering

dbt Labs