Developed Databricks catalog support within the Canner/WrenAI repository, enabling users to specify a catalog during connection setup. This work involved refactoring the backend connection logic in Python to accept a catalog parameter while maintaining compatibility with existing configurations. The developer focused on robust API integration and expanded unit testing to ensure the new feature did not introduce regressions or disrupt prior functionality. By applying a targeted fix, the implementation ensured the catalog parameter was consistently honored across different configurations, improving reliability for multi-catalog Databricks environments. The result was reduced setup friction and enhanced integration reliability for end users.