Learn to use LlamaIndex to build an advanced RAG pipeline that uses the KDB.AI vector database to retrieve relevant data via semantic search and metadata filtering. In this example, we will ingest two large financial publications, one from before the 2008 financial crisis, and one from after the crisis. LlamaIndex will be leveraged to load, chunk, and ingest the publications into KDB.AI. When a user asks a question, LlamaIndex will use KDB.AI to retrieve the relevant chunks of data, and then send that relevant data to an LLM to generate a response for the user.
This RAG pipeline will give users the ability to ask questions about financial regulations before the crisis, after the crisis, and compare regulations from before and after the crisis to understand what caused it and why it should not happen again.
Download the Jupyter Notebook at the GitHub Repository to try it today!