Data Pipelines — ScreeningA phone-screen round on orchestration fundamentals (Airflow & Dagster). Answer from first principles.

What does it mean for a pipeline task to be idempotent, and why does it matter?

What is a sensor in Airflow?

In Airflow, what does catchup=True do for a DAG with a start_date in the past?

The execution_date / data_interval of a scheduled run represents...

A task writes results with INSERT on each run. It gets retried and now has duplicate rows. What's the idempotent fix?

Why can a single stuck sensor stall many other DAGs on the same Airflow setup?