🤖 AI Agent Chatbot (FastAPI + Ollama + TinyLLaMA)

An interactive AI chatbot powered by Ollama and TinyLLaMA, served using FastAPI, and connected to a sleek custom frontend using HTML, CSS, and JavaScript.

---

🧠 Features

• 💬 Chat with a local LLM (tinyllama) using Ollama
• 🌐 FastAPI backend with RESTful API
• 📁 Static frontend served via FastAPI
• 💾 Persistent chat history saved in chatlog.json
• 🔁 Reset conversation functionality
• 📜 View conversation history via endpoint

---

🛠️ Tech Stack

• Backend: FastAPI
• Frontend: HTML + CSS + JS
• Model Runtime: Ollama
• Model Used: tinyllama (or any other compatible LLM via Ollama)
• Others: Pydantic, Requests, CORS, JSON

---

🔍 ML Concepts Applied:

• Inference pipelines
• Prompt logic
• Data persistence and session state

---

🚀 Getting Started

1. Clone the Repository

git clone https://github.com/yourusername/ai-agent-chatbot.git
cd ai-agent-chatbot

2. Setup Python Environment

python -m venv venv
venv\Scripts\activate     # On Windows
# source venv/bin/activate  # On macOS/Linux

pip install -r requirements.txt

3. Install Ollama and Model

Install Ollama for your OS.

Then run:

ollama run tinyllama

✅ Ensure Ollama is running in a separate terminal before starting FastAPI!

---

4. Run the Backend

uvicorn main:app --reload

The server starts at: http://localhost:8000

📝 Sample Chat Log

This project includes a chatlog-demo.json file to demonstrate the structure of saved conversations.
Note: Real-time chat logs are stored in chatlog.json, which is excluded from version control to protect user privacy.

---

5. Access the Frontend

Open your browser and visit:

http://localhost:8000/frontend/

You're ready to chat with your local AI! 🎉

---

📂 Project Structure

AI-agent-chatbot/
├── frontend/
│   ├── index.html
│   └── script.js
├── main.py
├── chatlog.json
├── requirements.txt
└── README.md

---

📡 API Endpoints

Method	Endpoint	Description
POST	/api/chat	Send a message to the chatbot
POST	/api/reset	Clear chat history
GET	/api/history	Get full chat history

---

🧹 Resetting Chat

Click the "Reset" button in the UI or call:

curl -X POST http://localhost:8000/api/reset

---

🔧 Future Enhancements

• Add multi-agent orchestration logic with role-based LLMs (e.g., Research Agent, Summarizer Agent)
• Integrate embeddings-based memory for personalized responses
• Lightweight UI with feedback metrics

---

🙋‍♀️ Made by

Shreya H S Machine Learning & Cloud Enthusiast 🌙☁️🤖 Do Checkout - Medium Post

---

📄 License

This project is open-source and free to use. Modify and build upon it as needed!