<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>OpenShell on Philip Brazeale</title><link>https://pbrazeale.github.io/tags/openshell/</link><description>Recent content in OpenShell on Philip Brazeale</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 05 Apr 2026 14:14:07 +0100</lastBuildDate><atom:link href="https://pbrazeale.github.io/tags/openshell/index.xml" rel="self" type="application/rss+xml"/><item><title>From Debian Laptop Notes to a Real AI Lab Stack</title><link>https://pbrazeale.github.io/posts/opencalw-build-notes-01/</link><pubDate>Sun, 05 Apr 2026 14:14:07 +0100</pubDate><guid>https://pbrazeale.github.io/posts/opencalw-build-notes-01/</guid><description>My experiment last week — NemoClaw Setup: Thinkpad T14s, Debian 13.4 — was a success. Not a polished success, not an enterprise success, but a success nonetheless.
It answered the question that mattered at the time: can I get OpenShell, OpenClaw, and NemoClaw running on real Linux hardware, outside the clean-room fantasy of vendor demos and idealized docs? Yes. And once that answer became yes, the next question showed up immediately: can I operate this stack sanely, repeatedly, and safely?</description></item><item><title>NemoClaw Setup: Thinkpad T14s, Debian 13.4</title><link>https://pbrazeale.github.io/posts/nemoclaw-setup/</link><pubDate>Fri, 27 Mar 2026 14:34:07 +0100</pubDate><guid>https://pbrazeale.github.io/posts/nemoclaw-setup/</guid><description>Prep Hardware Pick your device. I went with a T14s, running Debian 13 &amp;ldquo;Trixie&amp;rdquo; Accounts Setup your Nvidia account if you don&amp;rsquo;t have one. You&amp;rsquo;ll also need a build.nvidia.com account Generate your API key Setup OpenAI Keys (Or other provider) This will be handy as the agent swarm expands Docker https://docs.docker.com/engine/install/debian/ Run the following command to uninstall all conflicting packages:
sudo apt remove $(dpkg --get-selections docker.io docker-compose docker-doc podman-docker containerd runc | cut -f1) Install using the apt repository</description></item></channel></rss>