I built my own local AI assistant with Python (runs fully offline)

Published: (February 9, 2026 at 08:07 PM EST)
1 min read
Source: Dev.to

Source: Dev.to

Hi!

Over the last few weeks I built a small personal AI assistant just for learning.
It runs 100 % locally on my PC (no cloud).

Features

  • per user memory
  • voice replies (TTS)
  • conversational companion‑style chat
  • basic image understanding
  • language change according to user (ES/EN)

I wanted to understand how everything works under the hood instead of using APIs.

If anyone wants to test it and give feedback, I’d love to share the demo link 🙂

Demo: https://shortly-overmodest-dangelo.ngrok-free.dev/

0 views
Back to Blog

Related posts

Read more »

New article

Are you sure you want to hide this comment? It will become hidden in your post, but will still be visible via the comment's permalink. Hide child comments as we...

Build a Serverless RAG Engine for $0

Introduction: The Problem with “Toy” RAG Apps Most RAG tutorials skip the hard parts that actually matter in production: - No security model: Users can access...

Set up Ollama, NGROK, and LangChain

markdown !Breno A. V.https://media2.dev.to/dynamic/image/width=50,height=50,fit=cover,gravity=auto,format=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fu...