What is RAG? #
Retrieval-Augmented Generation (RAG) optimizes the output of a LLM, so it references a knowledge base outside its training data, before generating a response. RAG extends the capabilites of LLMs to a specific domain or source. with RAG we we can provide a custom knowledge base, ensuring the chatbot stays accurate and domain-specific, because it only references to the knowledge base we have uploaded.
My chatbot #
I used our schools internal curriculum, and that was the key knowledge base of our chatbot. I used Dify.ai to get started on building a simple chatbot, that could answer questions relevant to our curriculum.