04 Oct 2026
I cloned my own voice with open models on my laptop, so my deck engine can read a talk's script in my voice. The licence ruled out several models before I'd heard any of them, and for my voice, getting the ten-second sample right mattered at least as much as which model I used.
14 Sep 2026
Six principles for trusting a knowledge base an AI reads and writes in, and a starter kit that puts them in place. The repo is public from today.
#AI#Claude-Code#Second-Brain#Knowledge-Management#Open-Source
25 Jul 2026
Every second-brain system solves capture. Almost none solve trust. Six principles that spell GOVERN, and the three steps that put them in place, starting from an empty folder.
19 Jul 2026
I asked a close collaborator what people get most wrong about AI. The answers kept circling back to agreements, honesty labels and the restart button.
12 Jun 2026
An agent harness is a single while-loop that calls a model and dispatches tools - the only durable response to it is reading what's inside the box.
21 Feb 2026
How to call or text your OpenClaw AI bot using Vonage APIs. This guide is unofficial and based on my experience with these technologies.