During this course, you have learned about the fundamentals of LLMs, how to interact with them using langchain, and how Neo4j can support the LLM in answering questions.
In this optional challenge you will apply this new knowledge and skills to a new problem.
You will create a conversational agent using Langchain that can answer questions about a topic you are interested in.
The agent should:
-
Use a prompt that sets the scene for the LLM
-
Be given additional context that is relevant to the scenario
-
Have conversational memory and be able to answer a string of questions
-
Be given a few-shot example to support it in answering a specific type of question
I recommend that you approach this challenge in the following order:
-
Choose a topic that you are interested in
-
Create a prompt and create a chain that uses the prompt
-
Create an agent and add conversational memory
-
Add additional context to the agent
-
Add a few-shot example to the agent
Next Steps
You can continue learning about Neo4j, LLMs and GenAI on the following GraphAcademy courses: * Introduction to Vector Indexes and Unstructured Data - Understand and search unstructured data using vector indexes * Build a Neo4j-backed Chatbot using Python - Get hands-on and create a chatbot with Neo4j, Python, and Streamlit
You can learn more about Vector Indexes and Unstructured creating chat bots using Neo4j in the next GraphAcademy course - Build a Neo4j-backed Chatbot using Python.
Summary
In this optional challenge, you used the knowledge and skills you have learned during this course to create a conversational agent.
Congratulations on completing this course. I hope you have enjoyed it and learned a lot.