Elementary provides several features that align with your use case, but there are some limitations specific to Redshift that you should be aware of. ### Comparing Dev and Production Models Elementary supports comparing tables to flag data differences, which is useful for understanding the impact of changes between environments (e.g., Dev vs. Production). This functionality is available through Python tests for BigQuery and Snowflake, where you can compare two tables and identify differences. However, the knowledge sources do not confirm that this feature is currently available for Redshift. For Redshift, Elementary’s unstructured data validation and LLM-based tests are not yet supported, and documentation for such integrations is still in progress. You may want to reach out to the Elementary team for updates on Redshift support for this specific feature[Redshift support](https://docs.elementary-data.com/data-tests/ai-data-tests/supported-platforms/redshift). ### Anomaly Detection and Data Quality Monitoring Elementary excels at monitoring for anomalies and unexpected behavioral changes in your data models. You can configure anomaly detection tests (e.g., for volume, freshness, null rates, and column-level anomalies) in your dbt project. These tests compare recent metrics to historical values and flag significant deviations, which can help developers identify both expected and unexpected changes introduced during development. These tests are configured and executed like any other dbt test, and results are available in the Elementary UI and alerting system. This approach works across supported warehouses, including Redshift, for standard anomaly detection tests[How anomaly detection works](https://docs.elementary-data.com/data-tests/how-anomaly-detection-works) [Add anomaly detection tests](https://docs.elementary-data.com/data-tests/add-elementary-tests). ### Summary - Direct table comparison (Dev vs. Prod): Supported for BigQuery and Snowflake via Python tests, but not currently documented for Redshift. - Anomaly detection and monitoring: Fully supported for Redshift, allowing developers to detect unexpected changes in their models during development and testing. If your primary need is anomaly detection and monitoring for unexpected changes, Elementary can support this in Redshift. For direct table comparison, you may need to wait for future support or contact the Elementary team for more details. Let me know if you need more specifics or guidance on configuring these tests!