Overview

This course is for AI engineers, application developers, data engineers, data analysts, other data professionals, and students who want to build autonomous AI agents that work with enterprise data. Whether you're looking to move beyond basic chatbots, create intelligent systems that can query both structured databases and unstructured documents, or integrate AI capabilities into existing business workflows, this course provides the hands-on foundation you need. By the end of this course, you will be able to: - Distinguish between AI assistants and AI agents, and identify when autonomous agent capabilities are the right solution for your use case - Build Cortex Analyst and Cortex Search that enable agents to query structured metrics and search unstructured content using natural language - Configure and deploy Cortex Agents that autonomously plan tasks, select appropriate tools, and synthesize insights from multiple data sources - Write effective orchestration instructions that guide agent behavior and optimize response quality - Evaluate agent reliability using observability features and implement improvements based on performance metrics - Connect agents to external applications using Model Context Protocol for broader integration To be successful in this course, you should have basic familiarity with SQL and understand how data is organized in databases. Prior experience with AI or machine learning is helpful but not required, as the course covers foundational agent concepts before moving into implementation. This is a hands-on course where you'll build alongside the instructor, so you'll use a free Snowflake trial account.

Syllabus

Foundations of Agentic AI with Snowflake Cortex

In this module, you'll learn what makes AI agents different from traditional AI assistants and why this distinction matters for enterprise applications. You'll explore how agents autonomously plan, execute, and reflect on complex tasks rather than simply responding to direct queries. The module covers Snowflake's agentic AI architecture including Cortex Agents, Cortex Analyst, and Cortex Search, and how these components work together. You'll examine real-world use cases in B2B sales intelligence and customer service to understand when agent capabilities are the right solution for business problems.

Building Your First Cortex Agent

In this module, you'll build a complete B2B sales intelligence agent from scratch using Snowflake's visual interfaces. You'll create a semantic view for Cortex Analyst that enables natural language queries over structured sales metrics, configure Cortex Search for unstructured conversation transcripts, and connect both as tools to an agent. You'll write orchestration instructions that guide agent behavior and test your agent with questions that require structured data, unstructured data, and both combined. By the end, you'll have a working agent that autonomously decides which tools to use based on the question asked.

Advanced Agent Capabilities

In this module, you'll expand your agent skills with advanced capabilities for production use. You'll learn how agents handle complex multi-step scenarios through task decomposition. The module covers response optimization techniques including instruction refinement for orchestration and response format. You'll learn how to evaluate agent reliability using Snowflake's observability features. Finally, you'll configure Model Context Protocol to connect your agents with external applications, extending their reach beyond Snowflake.