Accelerating analytics with AI
Use an AI-assisted workflow to implement a small dbt feature end-to-end, then create a simple custom dbt agent skill to reinforce the workflow. You’ll draft a change (model + tests + docs), review it with a structured checklist for correctness, efficiency, and maintainability, and turn the key best practices into a reusable skill. The goal is safe acceleration: moving faster without lowering engineering quality.
After this lab, you will be able to:
- Use AI assistance to draft a dbt change (model + tests + docs) and iterate with intent
- Apply a practical review checklist to validate correctness, performance, and maintainability
- Produce a ready-to-merge change that meets team standards for naming, documentation, and testing
- Create a simple custom dbt agent skill that captures and reinforces core best practices for future work
- Comfortable reading SQL
Prerequisites:
- dbt Fundamentals
- Comfortable reading SQL
What to bring: You must bring your own laptop to complete the hands-on exercises. We will provide any required sandbox environments for dbt and the data platform.
Duration: 90 minutes
Check out more sessions
- Breakout session
ADBC in production: zero-copy, non-zero problems
Matt Topol / ColumnarView session - Hands-on lab
Beyond the basics: Running dbt at scale on Microsoft Fabric
Abhishek Narain / MicrosoftPradeep Srikakolapu / MicrosoftView session - Breakout session
The principals' framework for consistent dbt at scale
Thomas Antonakis / Kaizen GamingStefanos Nikolaou / Kaizen Gaming1 more speakerView session
