RAG chatbot demo: ask a dental clinic's documents anything

What is a RAG chatbot?

A RAG (retrieval-augmented generation) chatbot answers questions from a business's own documents instead of from general internet knowledge. It first retrieves the passages most relevant to the question, then a language model writes the answer from those passages and shows where it came from. This keeps answers accurate, up to date and checkable.

How this demo works

1. The fictional Lake Sarah Dental has four documents: a patient FAQ, a price list, office policies and an aftercare guide, split into 28 short passages.
2. On your first question, a small open-source embedding model (all-MiniLM-L6-v2, about 23 MB) loads into your browser.
3. The model turns every passage, and your question, into numbers that capture meaning, so "Are you open on Saturdays?" matches the opening-hours passage even without the same words.
4. It shows the closest passage, highlights the best sentence and names the source document. If the best match scores below 35%, it says the answer is not in the documents instead of guessing.

This demo shows the retrieval step. In the RAG chatbots we build for clients, a language model then writes a natural reply from the passages it finds.

Where businesses use RAG assistants

Common uses are answering customer questions on a website, helping staff find policies and procedures, and supporting support desks in clinics, online stores and professional services.