From the course: Enterprise AI Solutions with AWS: Amazon Q Business, Bedrock Knowledge Bases, and SageMaker MLOps

Unlock this course with a free trial

Join today to access over 24,900 courses taught by industry experts.

RAG workflows in Bedrock

RAG workflows in Bedrock

From the course: Enterprise AI Solutions with AWS: Amazon Q Business, Bedrock Knowledge Bases, and SageMaker MLOps

Start my 1-month free trial Buy for my team

RAG workflows in Bedrock

“

- [Instructor] Let's explore how you could enhance Amazon Bedrock with knowledge retrieval. In this case, we're going to talk through RAG or retrieval augmented generative AI. It's a fancy word, but really, all it means is that look specifically in this particular set of data to get your answers, but use the fancy generative AI techniques. So how do we do this? First step, we have the knowledge base, which is the blue section, and here is where we would put the sources. So it could be the Rust official documentation, it could be Crates.io Data, it could be the Standard Library, and all those would go into S3 and be ready for retrieval. Then we have the vector processing in this teal section here, and the Titan Embeddings would then take it and convert it into a searchable vector. And this makes the context much simpler for our later RAG process. And then in the system prompt here, this is where we would put in our expert mode configuration. So we would set the personality and say…

Contents