<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Writing — Raphael Lugmayr</title><link>https://raphaellugmayr.at/en/writing</link><description>Writing — Raphael Lugmayr</description><language>en</language><item><title>Evaluating AI agents: from a good impression to a useful result</title><link>https://raphaellugmayr.at/en/writing/ki-agenten-pruefen</link><guid isPermaLink="true">https://raphaellugmayr.at/en/writing/ki-agenten-pruefen</guid><description>An agent can report convincingly while doing the wrong work. A framework for checking tasks, results, permissions and repeatable trials.</description><pubDate>Fri, 11 Sep 2026 00:00:00 GMT</pubDate></item><item><title>AI agents beyond the chat window</title><link>https://raphaellugmayr.at/en/writing/ki-agenten-statt-chatfenster</link><guid isPermaLink="true">https://raphaellugmayr.at/en/writing/ki-agenten-statt-chatfenster</guid><description>A chat window can help with ideas, but workflows need more. Why operational AI needs a workspace and a person to approve its actions.</description><pubDate>Thu, 25 Jun 2026 00:00:00 GMT</pubDate></item></channel></rss>