AI Assignment Generator

Live Demo: https://assignmentcreatorte5et.vercel.app/

Demo Video: https://drive.google.com/file/d/1WkQXU9whEcAfdaBnaeTykMEl16EI23s9/view?usp=sharing

An AI-powered Assignment Generator that creates structured academic assignments dynamically using AI.

---

Features

• AI-generated assignment papers
• Structured question generation
• Real-time updates using WebSockets
• Background job processing with BullMQ
• PDF/file uploads using Cloudinary
• Assignment history & detail pages
• Responsive UI
• Proper form validation
• Zustand state management

---

Tech Stack

Frontend

• React.js
• Next.js
• TypeScript
• Tailwind CSS
• Zustand
• Socket.IO Client

Backend

• Node.js
• Next.js API Routes
• TypeScript

Database & Queue

• MongoDB
• Redis (Upstash)
• BullMQ

Services

• Gemini API
• Cloudinary
• Socket.IO

---

Architecture Flow

Frontend (Next.js)
        │
        ▼
API Route (/api/assignment/create)
        │
        ▼
BullMQ Queue (Redis / Upstash)
        │
        ▼
Worker Server (Render)
        │
        ▼
Gemini API
        │
        ▼
MongoDB
        │
        ▼
Socket.IO Update
        │
        ▼
Frontend

---

Project Structure

app/
│
├── api/
│   └── assignment/
│       ├── create/
│       │   └── route.ts
│       ├── [id]/
│       │   ├── route.ts
│       │   └── page.tsx
│       ├── assignments/
│       │   └── page.tsx
│       └── createassignment/
│           └── page.tsx
│
├── mobilebottomnav/
├── navbar/
├── sidebar/
├── components/
│
├── lib/
│   ├── buildprompt.ts
│   ├── bullredis.ts
│   ├── cloudinary.ts
│   ├── db.ts
│   ├── gemini.ts
│   ├── queue.ts
│   ├── redis.ts
│   ├── socket.ts
│   ├── utils.ts
│   └── worker.ts
│
├── socket-server.ts
│
├── store/
│   └── assignmentStore.ts
│
└── models/
    └── schema.ts

---

Assignment Generation Flow

1. User submits assignment request
2. API stores pending assignment
3. Job added to BullMQ queue
4. Worker processes assignment generation
5. Gemini generates structured questions
6. Result stored in MongoDB
7. Socket event updates frontend in real-time

---

AI Output

The system converts user input into a structured prompt and generates:

• Sections (A, B, C...)
• Questions
• Difficulty levels
• Marks distribution

LLM responses are structured before rendering.

---

Validation

• No empty fields
• No negative values
• File validation
• Structured request validation

---

Routes

Route	Description
/	Landing Page
/assignments	All Assignments
/assignment/[id]	Assignment Details
/createassignment	Create Assignment
/api/assignment/create	Create Assignment API
/api/assignment/[id]	Fetch Assignment API

---

Deployment

Service	Platform
Frontend	Vercel
Worker Server	Render
Socket Server	Render
Redis	Upstash
File Storage	Cloudinary

---

Environment Variables

MONGODB_URI=

REDIS_URL=

GEMINI_API_KEY=

NEXT_PUBLIC_SOCKET_URL=

NEXT_PUBLIC_CLOUDINARY_CLOUD_NAME=

NEXT_PUBLIC_CLOUDINARY_UPLOAD_PRESET=

---

Installation

git clone <repo-url>

npm install

npm run dev

Run Worker:

npm run worker

Run Socket Server:

npm run socket

---

Author

Samradhi Rathore

Full Stack Developer | MERN Stack Developer