PDF Q&A RAG Pipeline: LlamaIndex, OpenAI & Pinecone

This workflow builds a complete Retrieval-Augmented Generation (RAG) pipeline for PDF documents, ideal for structured content like insurance policies, legal docs, or compliance files. It triggers on new PDFs in a Google Drive folder, uploads them to LlamaIndex Cloud for accurate parsing into clean Markdown, then normalizes the text by stripping headers, footers, page numbers, and formatting artifacts using custom regex in a function node. Next, it intelligently splits the normalized content int