📚 PrepWise: Personalized AI Exam Assistant

PrepWise is your AI-powered study companion built for students preparing for competitive exams like NEET, JEE, UPSC, and more. It generates personalized learning plans, explains topics, answers doubts in real time, and creates quizzes — all powered by Agno, Mem0, and Groq.

🌟 Made for the Global Agent Hackathon – May 2025

---

🔥 Key Features

• 📖 Personalized Study Plans based on exam, subjects, and daily study time
• 🧠 Topic-wise Explanations with memory-driven chat support
• 💬 Ask Doubts Freely with context-aware answers using Groq
• 🧪 Quiz Generation with 10–20 MCQs in valid JSON format
• 📝 Agent Memory powered by Mem0 to track progress and goals
• 📱 Mobile & Web Support via React Native + Expo

---

🛠 Tech Stack

Layer	Technology
Frontend	React Native (Expo), Redux Toolkit
Backend	FastAPI
AI Agent	Agno, Groq, OpenAIChat
Tools	Firecrawl, Exa, ReasoningTools
Memory	Mem0
Deployment	Node.js, Python, MongoDB

---

📦 Monorepo Structure

prepwise/
├── frontend/
│   ├── app/
│   │   ├── (tabs)/
│   │   │   ├── _layout.tsx
│   │   │   ├── index.tsx
│   │   │   ├── profile.tsx
│   │   │   ├── quiz.tsx
│   │   ├── _layout.tsx
│   │   ├── +not-found.tsx
│   │   ├── setup.tsx
│   │   ├── welcome.tsx
│   ├── components/
│   │   ├── QuizCard.tsx
│   │   ├── ThemedText.tsx
│   │   ├── ThemedView.tsx
│   │   ├── TopicCard.tsx
│   ├── hooks/
│   │   ├── hooks.ts
│   │   ├── useColorScheme.ts
│   │   ├── useColorScheme.web.ts
│   │   ├── useThemeColor.ts
│   ├── store/
│   │   ├── slices/
│   │   │   ├── learningPlanSlice.ts
│   │   │   ├── userSlice.ts
│   │   ├── store.ts
│   ├── utils/
│   │   ├── api.ts
├── backend/
│   ├── app/
│   │   ├── __init__.py
│   │   ├── main.py
│   │   ├── __pycache__/
│   │   │   ├── main.cpython-313.pyc
│   │   ├── agent/
│   │   │   ├── __init__.py
│   │   │   ├── exam_agent.py
│   │   │   ├── profile_setup.py
│   │   │   ├── quiz_agent.py
│   │   │   ├── __pycache__/
│   │   │   │   ├── __init__.cpython-313.pyc
│   │   │   │   ├── exam_agent.cpython-313.pyc
│   │   ├── core/
│   │   │   ├── __init__.py
│   │   │   ├── config.py
│   │   ├── routes/
│   │   │   ├── __init__.py
│   │   │   ├── agent_routes.py
│   │   │   ├── __pycache__/
│   │   │   │   ├── __init__.cpython-313.pyc
│   │   │   │   ├── agent_routes.cpython-313.pyc
│   │   ├── utils/
│   │   │   ├── __init__.py
│   │   │   ├── logger.py

---

⚙️ Setup

Frontend (Expo React Native)

cd frontend
npm install
npm run start

Backend (FastAPI)

cd backend
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload

---

🧠 How It Works

1. Onboarding

• User profile saved in memory via profile_setup.py

2. Learning Plan

• exam_agent.py + generate_learning_plan() use memory to create subject-wise roadmap

3. Topic Explanation & Chat

• Explanations stored in Redux
• Chat allows doubt resolution using Groq

4. Quiz Generator

• quiz_agent.py fetches memory, generates valid JSON quiz
• Supports 10-20 MCQs with answers and explanations

---

📺 Demo Video

🎥 https://youtu.be/YOUR_VIDEO_LINK (Replace with your link)

---

👤 Team

GitHub: @finneysolomon

Role: Full Stack Engineer — Specialized in building full-stack mobile and web applications using React Native, MERN stack, Python (FastAPI), and AI agent frameworks like Agno. Skilled in creating end-to-end experiences from UI to intelligent backend agents.

---

---

📌 Notes

• Modular agent setup for learning plan, quiz, and profile
• Valid JSON responses for quiz and plan ensure robustness
• Built with future expandability (Socket.IO, voice input, etc.)

---

⭐️ Like it?

Star the repo and share with #GlobalAgentHackathon

---

PrepWise © 2025 — Made with ❤️ using Agno, Groq & Mem0