Do you have unstructured data sitting around that you can’t get value out of? Let’s turn it into something useful! Learn how to navigate pesky mixed-media pdfs and use few shot prompting with images to improve data accuracy.

This session will cover techniques used to convert problematic unstructured documents into structured data you can look at in aggregate and use to drive decision making. We’ll cover a few different real-world document types, and a step by step approach of how each technique solves certain problems. After we build up a local example and cover questions, we will look at what a scaled cloud native version of a solution like this might look like, as well as cases it could be applied to.

 We will use techniques like:

 – Multi modal models
 – Image based few shot prompting
 – Structured LLM outputs
 – Evaluations