dbt-cortex-pipeline

To build a dbt pipeline on Snowflake with Cortex AI, you can use an automated pipeline builder to configure data models, Semantic Views, Cortex Search, and Cortex Agents. This integrates text-to-SQL, document search, and conversational AI into a single deployable dbt project.

Semantic Views enable text-to-SQL natural language querying, while Cortex Search provides vector search over unstructured text within a Snowflake dbt project. Both are Cortex AI services that integrate with your data models to enable advanced AI interactions.

Yes, you need both dbt and Snowflake to construct end-to-end pipelines integrating Cortex AI services. The pipeline requires these environments to handle data modeling, orchestration, and deployment across bronze, silver, and gold data layers.

Yes, dbt supports bronze, silver, and gold data layers with various capabilities including pass-through models and AI enrichment. This structure allows you to progress from raw data to business-ready fact tables integrated with Cortex AI services.

The best way to add conversational AI to Snowflake data models is by integrating Cortex Agents into your dbt pipeline. This deploys conversational AI interactions directly alongside your structured data models and vector search capabilities.