Building Agentic RAG With Llamaindex L3 Building An Agent Reasoning Loop

This notebook ports the DeepLearning.AI short course Building Agentic RAG with Llamaindex Lesson 3 Building an Agent Reasoning Loop to using Llama 3. It shows how to define a complete agent reasoning loop to reason over tools and multiple steps on a complex question the user asks about a single document while maintaining memory.

You should take the course before or after going through this notebook to have a deeper understanding.

Load the data

Setup the Query Tools

Setup Llama and Agent

Note: The LlamaIndex's FunctionCallingAgentWorker API doesn't work correctly with Groq Llama, so we use ReActAgent here.

Lower-Level: Debuggability and Control

Note: The LlamaIndex's FunctionCallingAgentWorker API doesn't work correctly with Groq Llama, so we use ReActAgent here.