Apache Druid 28.0.0 is a significant release for the open-source distributed database, featuring over 420 commits and contributions from 57 developers, aimed at enhancing real-time analytics. Key updates include improved SQL compliance and the graduation of async queries, deep storage queries, and UNNEST to general availability, with added support for UNION ALL and experimental window functions. The release also introduces default nested column ingestion, multi-topic Kafka ingestion, and experimental concurrent append and replace features to manage late-arriving data more efficiently. Additionally, it offers enhanced support for Azure BLOB storage and SQL-based array ingestion, along with upgrades to the Apache Calcite library for better ANSI SQL support. These enhancements are designed to improve query stability and performance, providing a more flexible querying experience and enabling more efficient data management. Users are encouraged to consult the upgrade notes due to some non-backward-compatible changes and to share feedback via the Druid Slack channel.