Enhancing Foundation Models with Your Data

This is a Practical Workshop on Amazon Bedrock and RAG (Retrieval-augmented Generation).

Should I use an existing foundation model (FM) or train a new model? Often, the answer will be 'use an existing FM' because these models are so capable and rapidly evolving in their generalized knowledge and capabilities. Plus, you can customize either the responses of the model or the model itself with your own enterprise data. We'll walk through Amazon Bedrock, which offers a choice of high-performing foundation models, and show how you can customize those FMs with your own data. In this hands-on session, I will guide you through how you can use any foundational model and perform pre-training/continuous training using Bedrock and implement Retriever-Answer Generator (RAG) using Aurora as a vector store in Python.