Replay Seattle 2024

Retrieval-Augmented Generation (RAG) using vector databases has become a standard in GenAI applications. You get the most accurate and up to date results from a language model when you provide it with the right context. Your data pipeline is as important as your language model.
In this talk we'll explore how RAG works in practice, from chunking, vectorizing and storing data in Astra DB to feeding a model context, and the benefits that durable execution brings to workflows like these. You'll learn how the Temporal Platform helps look after the most important aspect of a generative AI application, your data, and the building blocks you need to bring GenAI to production.