Explore the challenges and solutions for building production-ready Retrieval-Augmented Generation (RAG) systems over complex documents in this comprehensive conference talk. Delve into the intricacies of handling large-scale, messy data sources like PDFs with embedded tables. Learn about implementing effective parsing strategies for complex documents with embedded objects, and discover advanced indexing techniques that go beyond simple chunking. Examine various cutting-edge retrieval algorithms designed to handle queries about both tabular and unstructured data, weighing their use cases and trade-offs. Gain valuable insights from Jerry Liu, Co-founder and CEO of LlamaIndex, as he guides you through the entire process of creating a robust RAG pipeline capable of processing and leveraging complicated document structures in real-world applications.