RAG Bash a
RAG chatbot that uses admin-uploaded PDFs as a knowledge base to deliver contextual, streaming responses.
TypeScriptProject
A retrieval-augmented generation chatbot where admins upload PDF documents and users get answers grounded in that exact content. The system chunks and embeds documents into a vector store, retrieves relevant passages at query time, and injects them into the LLM prompt as context. Responses stream back token-by-token in real time using server-sent events.

The stack is TypeScript with a streaming API layer. PDFs are parsed and embedded at upload time, and similarity search finds the top-k relevant chunks per query. Everything is designed to demonstrate practical RAG without needing a massive vector database setup.