Smart Course Material Q&A Bot

A Streamlit + LangChain RAG app for asking questions about your course materials (PDF, DOCX, TXT) with multiple AI personalities.

---

Features

• Upload course material (PDF, DOCX, TXT)
• Ask questions and get answers with sources
• Choose from multiple AI personalities (Friendly Tutor, Strict Professor, etc.)
• Powered by Google Gemini (Generative AI) and LangChain

---

Quickstart

1. Clone this repo
2. Create a virtual environment (recommended)
3. Install dependencies:

   pip install -r requirements.txt

4. Set up your environment:
   • Copy .env.example to .env
   • Get a Google Gemini API key from Google AI Studio
   • Edit .env and set:
     • GOOGLE_API_KEY=your-key-here
     • (optional, but recommended for stability)
       • GOOGLE_EMBEDDING_MODEL=models/gemini-embedding-001
       • GOOGLE_CHAT_MODEL=gemini-2.5-flash

---

Running the App

Option 1: Main launcher

python main.py

Option 2: Direct Streamlit

streamlit run app.py

If port 8501 is busy, use another port:

streamlit run app.py --server.port 8502

---

Environment Variables

Variable	Purpose	Example Value
GOOGLE_API_KEY	Your Google Gemini API key	AIza...
GOOGLE_EMBEDDING_MODEL	Embedding model for vector search	models/gemini-embedding-001
GOOGLE_CHAT_MODEL	Chat model for answering questions	gemini-2.5-flash

---

Troubleshooting

• Embeddings error (404 or NOT_FOUND):
  • Set GOOGLE_EMBEDDING_MODEL=models/gemini-embedding-001 in your .env.
• Chat model error (404 or NOT_FOUND):
  • Set GOOGLE_CHAT_MODEL=gemini-2.5-flash in your .env.
• Port already in use:
  • Use --server.port 8502 or another free port.
• API key issues:
  • Make sure your key is valid and has Gemini API access enabled.

---

Project Structure

File / Folder	Purpose
app.py	Streamlit web UI
main.py	Launcher script
bot_core.py	Core RAG logic (loading, embeddings, chains)
personalities.py	AI personality definitions
requirements.txt	Python dependencies
.env.example	Example environment config
langchain_6_(rag).py	Optional CLI RAG demo (not used by app)

---

Credits

• Built with LangChain, Streamlit, and Google Gemini.

---

License

MIT