The most effective AI agents are only as good as the data pipelines feeding them. To move an agent from a “demo” to a production-ready asset, you need a robust, governed architecture that handles data transformation and enrichment at scale.

In this hands-on lab, you will step into the role of a data architect to build the backend of a Marketing & Sales Intelligence Agent. You’ll move beyond basic prompting to focus on the “upstream” engineering: transforming raw, unstructured feedback into high-signal datasets using Snowflake’s native AI and transformation engines.

WHAT TO EXPECT

In this workshop, you will move from raw data to a fully orchestrated AI agent by mastering the following Snowflake components:

• Automated AI Pipelines with Dynamic Tables: Architect “set-and-forget” pipelines that use Dynamic Tables to build declarative pipelines that automatically refresh—eliminating the need for complex ETL tools or manual scheduling.”
• Intelligent Enrichment with Cortex AI: Programmatically trigger Cortex AI functions (Sentiment Analysis, Classification, and Summarisation) directly within your SQL pipelines to enrich unstructured data at scale.
• The Semantic Layer: Construct a reliable Semantic Model that provides the context necessary for an agent to bridge the gap between business questions and technical schemas.
• Actionable Intelligence: Integrate structured sales data with enriched unstructured feedback to build a Marketing & Sales Intelligence Agent that identifies trends and churn risks.

Complete the hands-on lab and earn a Zero to Agent digital badge to share your achievement!

WHO SHOULD ATTEND
Data & AI Engineers, Architects, Analysts and Practitioners.

Prerequisites: 

• Bring a laptop + charger
• A Snowflake account (we’ll create a free trial account during the lab)
• Familiarity with Python and SQL
• A machine with access to download external files and repos from Github