“Databricks' Spark declarative pipelines now process 200 trillion rows of data every day.”