<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Field-Report on obuchmann.dev</title><link>https://obuchmann.dev/en/tags/field-report/</link><description>Recent content in Field-Report on obuchmann.dev</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 22 Oct 2026 00:00:00 +0200</lastBuildDate><atom:link href="https://obuchmann.dev/en/tags/field-report/index.xml" rel="self" type="application/rss+xml"/><item><title>The Factory's Biggest Customer</title><link>https://obuchmann.dev/en/posts/the-factorys-biggest-customer/</link><pubDate>Thu, 22 Oct 2026 00:00:00 +0200</pubDate><guid>https://obuchmann.dev/en/posts/the-factorys-biggest-customer/</guid><description>&lt;h2 id="shut-down-and-counted">Shut Down and Counted&lt;/h2>
&lt;p>On October 16, I shut down my first fully home-built AI factory. Before that, I exported all the relevant data and counted everything.&lt;/p>
&lt;p>The factory works tickets from a local Forgejo instance following a predefined SDLC, in its final form ten steps, from &lt;code>discover&lt;/code> through &lt;code>design&lt;/code> and &lt;code>impl&lt;/code> to the closing &lt;code>sentinel&lt;/code>. It tracks progress and decisions through labels and comments, communicates with me, and finally opens a PR for approval. The agents themselves ran in a sandbox, first locally in a container, two weeks later in a Kubernetes cluster.&lt;/p></description></item></channel></rss>