Building Your First Multimodal Document Pipeline

Dive hands-on into unstructured data and build your first intelligent document processing application from scratch. In this lab, you'll learn core implementation patterns by creating a pipeline that processes customer data across multiple formats: audio transcripts, PDF documents, images and structured tables. The lab progresses from basic data ingestion to more sophisticated workflows including entity mapping, content classification and pattern recognition across different data types.

In this lab, AI experts will provide step-by-step guidance on how to:
 

• Use familiar SQL commands to extract AI insights across multimodal data
• Build a working application that demonstrates key AI capabilities
• Integrate these features into your own projects using SQL and API patterns