Local AI Agent With RAG

Build a Local AI Agent with Python, using Ollama, LangChain, and RAG (Retrieval‑Augmented Generation).

This project lets you run a smart AI assistant entirely locally — no cloud API keys needed. It connects:

• a local LLM (Ollama)
• a vector store (Chroma)
• Python RAG workflow

so you can query over documents or datasets and get helpful answers.

---

🚀 Features

• 🧠 Uses local LLMs with Ollama for private inference
• 🔍 Adds RAG — AI can search and reason over your data
• 📚 Uses vector embeddings (via Chroma) for retrieval
• 🐍 Entirely Python‑based, easy to customize

---

📦 Requirements

You’ll need:

• Python 3.10+
• Ollama installed and running
  • Pull a local model (like a lightweight one)

    ollama pull # (model_name, e.x llama3.2:1b)

  • Pull a embed model (like a lightweight one)

    ollama pull # (model_name, e.x mxbai-embed-large)

• Virtual environment (recommended)

---

🛠 Installation

1. Clone the repo:

   git clone https://github.com/miuceo/local-rag-agent.git
   cd local-rag-agent