The ExamEdge Backend is a FastAPI-powered server designed to extract and process academic content from PDFs, generate AI-based summaries and MCQs, and recommend books, videos, and faculty based on content topics. This backend powers the ExamEdge app experience by delivering intelligent and interactive learning material.
- 📤 Upload and extract text + images from PDF documents
- ✨ Generate AI-powered summaries using Gemini
- ❓ Create MCQs chunk-by-chunk from document content
- 📚 Recommend relevant books using Google Books API
- 🎥 Fetch relevant YouTube videos per topic
- 🧑🏫 Suggest expert faculty members based on document topics
- 💬 RAG-based chatbot to interact with uploaded PDF content
- 🌍 Deployed on an NGINX server hosted on an Ubuntu EC2 instance
Upload a .pdf file, extract its content, and generate structured JSON metadata.
Returns:
- Original filename
- Unique filename
- Confirmation message
Generates a detailed summary of the uploaded PDF using Gemini.
Requires: Metadata to already exist from upload step.
Generates Multiple Choice Questions based on document content, starting optionally from a given page.
Returns one book recommendation per topic based on the PDF summary.
Returns one YouTube video per topic using YouTube API, based on document summary.
Returns matched faculty members using static data and document summary topics.
Provides an AI chatbot experience with the uploaded PDF. Uses document context to respond to user queries (RAG-based).
Request Body:
{
"query": "What is the main idea of chapter 3?"
}Returns: AI-generated response, Pages used for generating the response