I recently wrote a tutorial for freeCodeCamp on building a private RAG Q&A agent for your own documents using LangChain v1, Ollama, Qwen, and Python.

This is the kind of project I keep coming back to because it feels genuinely useful. I already have a lot of material sitting around like notes, papers, meeting summaries, random reference docs but most of it is only useful if I remember where it is and take the time to dig through it. What I really want is to just ask a question in plain English and get back an answer grounded in my own documents.

What makes this setup especially appealing to me is that it runs locally. That means the documents stay on my machine, I’m not paying per-call API costs, and I’m not turning private notes into data for someone else’s service. It feels like a much more practical and personal way to use AI.

Writing this tutorial was a lot of fun, and it pushed me further in a direction I already feel strongly about: local, on-device AI agents are going to matter a lot. I think there’s something powerful about tools that work with your own data, on your own machine, in a way that feels private and under your control.

You can read the full article here:

How to Build a RAG Q&A AI Agent for Your Documents Using LangChain v1.