Smart Assistant for Document Summarization, QA & Quiz Evaluation

An AI-powered assistant built with Django and Groq LLM that can:

• Auto-summarize uploaded documents (≤150 words)
• Answer questions based on document context
• Provide one-line justified answers
• Generate quizzes and evaluate your answers with scores

---

Features

• Upload PDF/TXT documents
• Auto-summary using Groq (DeepSeek R1 Distill LLaMA 70B)
• Ask Anything mode using context-aware retrieval
• FAISS vector index on document chunks
• Context retrieval with MiniLM sentence embeddings
• Quiz Mode: 3 questions generated from your uploaded document
• Evaluation with:
  • Score (0–100)
  • Justification
  • Correct Answer
• Clean Bootstrap UI
• Reset chat/quiz anytime

---

🛠️ Tech Stack

Layer	Tools Used
Backend	Python, Django
Frontend	HTML, CSS, Bootstrap
Embeddings	all-MiniLM-L6-v2 via HuggingFace
LLM (QA+Summary)	deepseek-r1-distill-llama-70b via Groq API
Quiz + Evaluation	LangChain + Prompt Engineering
Vector Search	FAISS
Document Parsing	PyMuPDF

---

Project Structure

smart-assistant/
├── assistant/
│ ├── templates/
│ │ ├── upload.html
│ │ ├── summary.html
│ │ ├── ask.html
│ │ └── quiz.html
│ ├── utils/
│ │ └── text_processing.py
│ ├── views.py
│ ├── models.py
│ └── urls.py
├── vectorstore/
│ └── db_faiss/
├── media/
│ └── (Uploaded PDFs)
├── .env
├── .gitignore
├── manage.py
├── requirements.txt
└── README.md 

---

Local Setup Instructions

1. Clone the Repository

git clone https://github.com/singhadi01/smart-assistant.git
cd smart-assistant

2. Create and Activate Virtual Environment

python -m venv env
env\Scripts\activate     # For Windows
# OR
source env/bin/activate  # For macOS/Linux

3. Install Dependencies

pip install -r requirements.txt

4. Add your Groq API Key

GROQ_API_KEY=your_groq_api_key_here

4. Run Django Server

python manage.py runserver

---

Screenshots

---

Author

Aditya Singh
Email: singhaditya270305@gmail.com
GitHub: singhadi01 © 2025 — Smart Assistant Project