RAGz
Upload documents and ask questions about them in a chat window.
- Status
- Prototype · hosted demo unavailable
- Year
- 2025
- Role
- Collaborative
- Area
- Agents & AI systems
In plain English
Hunting through a pile of PDFs for one answer is slow. RAGz lets you upload documents and ask questions, and it answers from the most relevant passages. The technique is called retrieval-augmented generation.
How it works
- Upload PDF documents through the web page.
- Split them into passages and store them in a vector database (a search index based on meaning).
- For each question, find the closest passages and rerank them (reorder them by relevance).
- A language model writes the answer from those passages, and the chat remembers your session.
What I did
I built it together with collaborators. The browser chat and session handling are part of the code in my repository.
What I learned
Search quality decides whether a document chatbot is useful, so getting the right passages in front of the model is the heart of the work.