Local AI Agent with RAG

A powerful local AI assistant that utilizes Retrieval-Augmented Generation to answer user queries based on custom documents using local LLMs and vector databases.

🌟 Features

Local document ingestion

Vector embedding and retrieval using FAISS

Query handling through LangChain and LLMs

Custom knowledge base

⚡ Tech Stack

Python

LangChain

FAISS

Local LLM (GPT4All or similar)

Streamlit (optional frontend)

📆 Use Cases

Custom chatbot trained on PDFs or markdown files

Personal knowledge assistant

FAQ bot for local documentation

🔧 Setup

pip install -r requirements.txt python ingest.py # to process your documents python app.py # to launch the assistant

🚀 How It Works

Ingest: Parses and converts documents into vector embeddings

Store: Saves vectors in FAISS index

Query: Embeddings used to retrieve top results

Generate: Local LLM answers based on retrieved content

🌐 Demo

Try it on your terminal or deploy via Streamlit for a web-based version.