FIVETRAN AND DBT LABS, TOGETHER / SNOWFLAKE HOL

Your personal dbt agent wherever you work

Our first joint product launch as Fivetran + dbt Labs. Built on Snowflake. Public beta on June 1, 2026.

A 20-minute hands-on lab where you'll use dbt Wizard to investigate a real retail dataset, extend an existing model with a new source, and fix a broken pipeline - all without writing SQL by hand or breaking downstream models.

Fivetran dbt Labs Snowflake
See it in action

The onboarding scenario, end-to-end

Watch a working example before you walk in the door. You'll do this one first on Monday.

Watch the onboarding scenario play through end-to-end. The other four scenarios will be linked individually below.

The product

What is dbt Wizard?

dbt Wizard is an AI agent built from the ground up for the way analytics engineers actually work. Not just code generation, but the entire data lifecycle: asking questions, investigating, understanding, changing, validating, and shipping. It's grounded in real dbt project context: lineage, tests, contracts, and defined metrics. And it's available wherever analytics engineering work actually happens.

dbt Wizard

Chat-first workspace

A conversational workspace in the dbt platform, plus an embedded assistant inside dbt Studio. Tailored capabilities ship in dbt Canvas and dbt Insights for visual modeling and analytics work.

dbt Wizard CLI

Terminal-native agent

A terminal-native agent for local development. Runs against your dbt project on your laptop, with your warehouse credentials, in the editor you already use. This is what the HOL uses.

Why dbt Wizard

What changes when you use it

The shift isn't "AI writes SQL for you." It's that the agent understands your project, respects your lineage, and validates its own work before handing it back.

Without dbt Wizard

  • Context switching
  • Risky changes that ignore lineage
  • Inconsistent validation
  • No guardrails for non-technical users
  • DIY tax configuring MCP servers
  • Slower delivery
  • Increased production incidents
  • Ungoverned risk
  • Engineering bottlenecks
  • AI ROI that never materializes

With dbt Wizard

  • No more context switching
  • Validation by default
  • Governed by default
  • Controlled access for non-technical users
  • Zero DIY tax
  • Faster time to ship
  • Safer changes
  • Higher trust
  • Improved engineering capacity
  • AI ROI that materializes
Capabilities

What dbt Wizard actually does

Ten capabilities that distinguish a purpose-built AE agent from a generic code assistant.

01

Full analytics loop without context switching

Investigate, change, validate, and ship from one surface.

02

Grounded in your full dbt project

Lineage, contracts, semantic definitions, tests, governance - all in context.

03

Native dbt metadata engine

Reads your project the way dbt itself reads it. No translation layer.

04

Coordinated, multi-file changes

Edits across models, schema files, and tests in a single coherent change.

05

Best practices built in

A decade of analytics-engineering knowledge maintained by dbt Labs and the community.

06

Validates its own work

Built-in comparison loop runs your changes against expected outputs before handoff.

07

Human in the loop by default

You stay in control. The agent proposes; you approve.

08

Full transparency

Reasoning and tool calls are visible. Nothing happens in a black box.

09

Specialized agent harness

Purpose-built for AE workflows, not a generic chatbot in a costume.

10

Warehouse-agnostic

Snowflake, BigQuery, Databricks, Redshift, and others.

What you'll do onsite

Five scenarios. One lab.

Each scenario is a self-contained workflow that takes roughly 10-15 minutes. Start with onboarding, then pick whichever scenarios match your role. The choose-your-path scenarios let you pick from three real business situations.

0
Start here

Onboarding

New AE at The Builder Depot - get a tour, understand the project, ship a first model.

7 steps
1
Scenario 1

Inventory Misallocation

Find which stores got the wrong stock allocation before the big sale.

6 steps
2
Scenario 2

Extending an Existing Model with a New Source

Choose-your-path x 3: Customer 360, Operations, or Merchandising.

6 steps 3 paths
3
Scenario 3

Marketing Targeted Campaigns

Build a 180-day customer activity layer and segment VIPs, big spenders, and category-loyal customers.

7 steps
4
Scenario 4

Broken Model from a Source Column Rename

Reproduce a failure, find the blast radius, fix the alias chain, re-run to green. Choose-your-path x 3.

5 steps + terminal 3 paths
Before you arrive

What you'll need

Get this done before Monday morning so we don't burn lab time on installs.
Early results

What beta users are saying

Customer quote

"What really sets the dbt Developer Agent apart is its precision in identifying and resolving bottlenecks. It doesn't just suggest code; it understands our existing tests and lineage well enough to troubleshoot issues almost instantly. This has significantly reduced our build times and allowed us to scale our dbt project with total confidence in our data's trustworthiness."

Customer quote

"We went from about 60 conformance errors to 7, using the Fusion migration agent. That's the difference between too hard and actually doable."

Resources

Quick links

Install

Getting Started

Install dbt Wizard, clone the repo, set up your venv, validate your environment.

Printable

Prompt Sheet

A one-page printable cheat sheet of the exact prompts to use during the lab.

Snowflake

Infrastructure

How the warehouse, schemas, and per-attendee credentials are wired up.

Source

GitHub repo

The full dbt project, seed data, scripts, and lab assets used for the HOL.