Our dbt project hit 300 models. Things that kept it sane:
Reyansh Iyer
Data Engineer | Spark, Airflow, dbt, Snowflake | Pipelines that don't break on Mondays
Our dbt project hit 300 models. Things that kept it sane:
• Strict layers: staging → intermediate → marts
• Every model has an owner in its YAML
• Tests on every primary key (unique + not null) by default
• Slim CI: only run modified models + downstream
CI went from 40 min to 6 min with slim CI alone.
22 reactions2 comments
