Posts
All the articles I've posted.
-
Leveling Up the Edge of Local SLM AI
Here is how I build a better cage for the brain we have, transforming a simple Raspberry Pi 4B into a Sovereign Edge Agent with SLM, RAG, GBNF, QLoRA, MCP...
-
A Local-First, Self-Learning RAG AI on a RPi
This guide will walk you step-by-step through turning your Pi 4B into a dynamic AI knowledge base (RAG) that learns from your local folders and the web.
-
Traded SLM Intelligence for Raw Speed
I’ll show you how I stopped treating my Pi like a walking encyclopedia and turned a tiny 0.8B model into a high-speed "Logic Router" that powers my local agent infrastructure without the bloat.
-
Running thePopeBot with Qwen 3.5 2B in RPi
Here is exactly how I squeezed every drop of performance out of the Pi 4's Cortex-A72 architecture to run the Qwen3.5 2B model as the "brain" for my local agent.