In this talk, "Cooking RAG like a Chef de Cuisine," we explore the art and science behind implementing Retrieval-Augmented Generation (RAG) systems with Large Language Models (LLMs), using a variety of tools and techniques. Like a chef crafting a gourmet dish, we'll carefully select and explore different approaches to building RAG pipelines. From traditional retrieval methods to advanced frameworks like LangChain, Haystack, and LlamaIndex, we'll highlight how to combine these ingredients effectively.

We'll take a closer look at tools such as AI Studio, Promptflow/Prompty, and Semantic Kernel, examining their role in creating and managing LLM-powered RAG systems. In addition, we will explore deployment strategies on Azure, discussing optimization, scalability, and monitoring for production-ready pipelines.

Through hands-on examples, we’ll also cover evaluation techniques to ensure your system delivers high-quality results, as well as deployment strategies that help scale RAG systems in real-world scenarios. Whether you're new to RAG or looking to refine your approach, this session will provide practical, actionable insights to help you develop and deploy robust RAG pipelines—like a true chef de cuisine!