This RAG (retrieval-augmented generation) assistant answers questions from a library of university PDF documents and shows the file name and page number behind each answer. It is my own project, not a client engagement.
The problem
Universities and businesses keep important information in long PDFs: handbooks, policies, procedures. People can’t find what they need, so they ask a colleague, or they guess. A generic AI chatbot doesn’t help either, because it answers from general knowledge, not from your documents.
What I built
- Search across all the PDFs. The assistant finds the passages that answer a question, across the whole library.
- Answers grounded in those passages. It writes its answer from what it found.
- Citations on every answer. Each answer lists the file and page it came from, so anyone can open the source and check.
Why citations matter
Citations turn a chatbot from “trust me” into “here’s where it says so”. Staff can verify an answer in seconds, and wrong or outdated documents become easy to spot.
Current limits
The assistant reads PDFs that contain real text. Scanned PDFs, which are images of pages, aren’t supported yet, because the pipeline doesn’t include OCR (optical character recognition). For a document library with many scans, adding OCR would be the first step.
Where this applies to your business
Any team with knowledge locked in documents: policies, product manuals, contracts, SOPs or training material.
Frequently asked questions
What is a RAG document assistant?
It is a chatbot that first searches your own documents for the passages that answer a question, then writes its answer from those passages instead of from general knowledge.
Why does it show citations?
Every answer lists the file name and page number it came from, so staff can check an answer in seconds and spot wrong or outdated documents.
Can it read scanned PDFs?
Not yet. It reads PDFs that contain real text. Scanned PDFs are images of pages, so a library with many scans would need OCR added first.
What kinds of documents can it work with?
Policies, handbooks, product manuals, contracts, SOPs and training material: any knowledge that is locked in long documents.
Try it free: take the 3-minute AI readiness quiz, see what missed calls cost you with the AI savings calculators, try the in-browser RAG chatbot demo or listen to the AI receptionist demo.
Related services: RAG chatbot development · AI chatbot development
Read more: What is a RAG chatbot? · Book a free consultation
AI/ML engineer building LLM agents, RAG systems and automation for growing businesses.
