🧠 AI Research Assistant

An interactive AI-powered assistant built with LangChain and Streamlit, designed to help you:

• 🔍 Ask questions about PDF documents
• 🧰 Use tools like a calculator, file tools, and web search
• ✍️ Summarize documents in natural language

---

🚀 Features

1. 📄 Document Q&A

Upload a PDF and ask natural language questions about its content. The system uses document chunking, embeddings, and retrieval-augmented generation (RAG) to answer accurately.

2. 🧰 Tool Agent

Interact with an AI agent that can:

• Perform calculations
• Search the web (DuckDuckGo)
• Interact with the local file system (via FileManagementToolkit)

3. ✍️ Summarize Documents

Generate concise summaries of entire PDF files using custom prompt templates and an LLM chain.

---

🧱 Project Structure

.
├── README.md
├── app
│   ├── agents
│   │   ├── __init__.py
│   │   └── tool_agent.py
│   ├── chains
│   │   ├── __init__.py
│   │   ├── qa_chain.py
│   │   └── summarizer_chain.py
│   ├── main.py                     # Main Streamlit app
│   └── utils
│       ├── __init__.py
│       ├── document_loader.py
│       └── vector_store.py
├── pyproject.toml
└── requirements.lock

---

🛠️ Installation

1. Clone the repo

git clone https://github.com/your-username/ai_research_assistant.git
cd ai_research_assistant

2. Install dependencies

pip install -r requirements.lock

3. Set environment variables
   Create a .env file or export your OpenAI API key:

export OPENAI_API_KEY=your-key-here

4. Run the app

streamlit run app/main.py

---

📦 Dependencies

• langchain
• langchain-community
• langchain-openai
• openai
• streamlit
• faiss-cpu or chromadb
• PyPDF2, pdfminer.six, or similar for PDF reading

---

💡 Future Enhancements

• Chat history with memory
• Support for multiple PDFs at once
• Export summaries and chat logs
• UI polish with user authentication

---

🙌 Acknowledgments

Thanks to the LangChain, OpenAI, and Streamlit communities for their powerful open-source tools.

---