Blog
Technical deep-dives on building reliable data platforms, from the team that builds them.
A metadata database transforms your data platform from a collection of disconnected pipelines into a system you can actually understand, debug, and trust. Here is how it works and why it matters.
Read moreWe use the medallion pattern on every data platform we build. Here is the exact dbt project structure, naming conventions, and SQL patterns we follow — and the mistakes we made before landing on them.
Read moreWe have operated Airflow for three years and migrated two clients to Dagster. Here is when we pick each, with real config examples from production deployments.
Read moreA real cost optimization engagement: the exact queries we ran to find waste, the warehouse changes we made, and the monitoring we set up to keep costs down.
Read moreAn idempotent pipeline produces the same result whether you run it once or five times. Here is how we implement this in practice with Airflow, dbt, and Snowflake — with the exact SQL patterns we use.
Read moreThe dbt test suite and Great Expectations checks we run on every pipeline. Includes the actual YAML configs, custom tests, and the alerting setup that pages us when something breaks.
Read moreIncremental models are dbt's most powerful feature and its biggest footgun. Here are the three strategies we use, when to pick each, and the late-data handling pattern that took us a year to get right.
Read more