The principals' framework for consistent dbt at scale

At Kaizen Gaming, analytics engineering grew to around 50 engineers across six domain teams, each with different levels of dbt adoption and their own ways of solving similar problems. As principal analytics engineers, we needed to improve consistency without turning central governance into a bottleneck. 

This talk shares three patterns that helped us get there. We moved company-wide foundational models into a centrally owned core repository. We packaged shared macros and tests into a custom internal package with automation to propagate standards downstream. And we managed dbt platform configuration through Terraform with CI guardrails scoped to affected projects only. Together, these changes improved developer experience, reduced duplicated effort, and gave domain teams more autonomy by making the right defaults easy to reach for. 

If you're managing dbt at scale across multiple teams, you'll leave with patterns you can adapt directly.