Marquez is an open-source data lineage platform released by WeWork and developed as part of the OpenLineage ecosystem. It helps organizations collect, store, and visualize metadata from data pipelines, supporting use cases such as data governance, data quality monitoring, and performance analytics. At its core, Marquez provides a real-time metadata server with an OpenLineage-compatible endpoint for gathering information from running jobs and applications.

As the OpenLineage reference implementation, its API server works with community integrations including Apache Airflow, Apache Spark, Apache Flink, dbt, and Dagster. Marquez also offers a web interface that presents a visual map of dependencies across a data ecosystem. Users can browse metadata, inspect job inputs and outputs, trace dataset lineage, and review execution details and performance metrics.

Its lineage API enables teams to traverse dependency trees and query context across datasets, pipelines, and orchestration platforms. This information can support automation for tasks such as backfills and root cause analysis, and can enrich data catalogs and data quality systems