CogniGen Backend

FastAPI backend for CogniGen, an adaptive RAG-based cognitive assessment platform. The backend handles authentication, document ingestion, book processing, retrieval, question generation, evaluation, and quiz workflows for the React frontend.

Highlights

• Retrieval-augmented question generation from uploaded learning material.
• Multi-hop reasoning prompts for higher-order assessment questions.
• Evaluation module for answer quality and question validation.
• FastAPI service structure with schemas, models, auth, and database modules.
• Designed to pair with the cognigen-frontend React application.

Tech Stack

• Python
• FastAPI
• SQLAlchemy
• RAG / LLM orchestration
• ChromaDB-style local retrieval store

Architecture

flowchart LR
    Frontend["React frontend"] -->|REST API| API["FastAPI backend"]
    API --> Auth["JWT authentication"]
    API --> Books["Book upload and processing"]
    API --> Quiz["Quiz generation and grading"]
    Auth --> DB[("PostgreSQL database")]
    Books --> Extract["ZIP/PDF content extraction"]
    Extract --> Embed["Gemini embeddings"]
    Embed --> Vector[("Local Chroma vector store")]
    Quiz --> Retrieve["RAG retrieval"]
    Retrieve --> Vector
    Retrieve --> LLM["Groq/Gemini question generation"]
    LLM --> Quiz
    Quiz --> DB

Loading

CogniGen is organized as a REST API that keeps user, book, quiz, and response metadata in PostgreSQL while storing searchable learning chunks in a local vector store. Uploaded course material is extracted, chunked, embedded, and then retrieved during question generation so the assessment flow can cite source chapters and pages.

Run Locally

pip install -r requirements.txt
copy .env.example .env
uvicorn app.main:app --reload

Update .env with your local PostgreSQL connection string and API keys before starting the server. Do not commit .env.

The API runs at http://localhost:8000 by default, and the interactive FastAPI docs are available at http://localhost:8000/docs.

Configuration

Variable	Purpose
DATABASE_URL	PostgreSQL connection string used by SQLAlchemy.
SECRET_KEY	JWT signing secret for access tokens.
GROQ_API_KEY	LLM provider key for chat, generation, and grading flows.
GEMINI_API_KEY	Google Gemini key for embeddings and evaluation support.

Core API Surface

Method	Endpoint	Purpose
POST	/api/v1/signup	Register a teacher/user account.
POST	/token	Issue a JWT access token.
POST	/api/v1/books/upload	Upload zipped learning material for processing.
GET	/api/v1/books	List uploaded books for the authenticated user.
POST	/api/v1/chat	Ask questions across selected uploaded books.
POST	/api/v1/generate-questions-from-book	Generate assessment questions from retrieved source context.
POST	/api/v1/quizzes	Create a quiz from generated questions.
GET	/api/v1/quizzes/{quiz_id}	Open a quiz for student access.
POST	/api/v1/quizzes/{quiz_id}/submit	Submit student quiz answers.
GET	/api/v1/quizzes/{quiz_id}/results	View quiz responses and scores.