🤖 Python AI Agent

This is a simple AI-powered research assistant I built using Groq + LangChain. The idea is straightforward — you ask a question, and the agent tries to give a clean, structured answer, even using the web when needed.

---

🚀 What it can do

• Answer general questions
• Search the web for up-to-date info
• Return responses in a structured format (not messy text)
• Keep things fast using Groq (LLaMA models)

---

🛠 Tech used

• Python
• LangChain
• Groq API
• Pydantic

---

📁 Project structure

python-ai-agent/
│
├── main.py
├── requirements.txt
├── .gitignore
└── .env (not included)

---

⚙️ How to run it

1. Clone the repo

git clone https://github.com/MAHADEVAN-007/python-ai-agent.git
cd python-ai-agent

2. Create virtual environment

python -m venv venv
venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Add your API key

Create a .env file and add:

GROQ_API_KEY=your_api_key_here

5. Run the project

python main.py

---

💡 Example

Input:

What is ransomware?

Output (simplified):

Topic: Ransomware

Summary:
A type of malware that locks or encrypts data and demands money to unlock it.

Sources:
- FBI
- Microsoft

---

🧠 How it works (simple version)

• You enter a query
• The agent decides if it needs to search the web
• It generates an answer
• The response is formatted into a clean structure

---

🔧 Things I can improve later

• Add memory (chat history)
• Build a web interface
• Add more tools (APIs, calculator, etc.)
• Improve response quality

---

👨‍💻 Author

Raman Mahadevan GitHub: https://github.com/MAHADEVAN-007

---

If you find this useful, feel free to ⭐ the repo.