▲Came across this piece from Discord on how they scaled dbt for petabyte-level workloads and a big engineering team. Some interesting ideas around environment isolation, faster compilation, guardrails, and CI/CD patterns that data and analytics engineers might find useful.
reply