<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Support - Tag - Simone Vellei - Blog</title><link>https://simonevellei.com/en/tags/support/</link><description>Support - Tag - Simone Vellei - Blog</description><generator>Hugo -- gohugo.io</generator><language>en</language><managingEditor>henomis@gmail.com (Simone Vellei)</managingEditor><webMaster>henomis@gmail.com (Simone Vellei)</webMaster><copyright>© 2025 Simone Vellei. All rights reserved.</copyright><lastBuildDate>Wed, 08 Apr 2026 17:01:39 +0200</lastBuildDate><atom:link href="https://simonevellei.com/en/tags/support/" rel="self" type="application/rss+xml"/><item><title>Build a support agent that routes itself</title><link>https://simonevellei.com/en/build-a-support-agent-that-routes-itself/</link><pubDate>Wed, 08 Apr 2026 17:01:39 +0200</pubDate><author>Simone Vellei</author><guid>https://simonevellei.com/en/build-a-support-agent-that-routes-itself/</guid><description><![CDATA[<div class="featured-image">
                <img src="/images/phero003.png" referrerpolicy="no-referrer">
            </div><p>A lot of support bots fail for the same reason. They try to do everything with one prompt.</p>
<p>Billing questions, outage reports, refund requests, API errors, onboarding questions. It all goes into one agent, and the prompt turns into a long list of rules, exceptions, and fallback behavior. At first it feels convenient. A few weeks later it feels fragile.</p>
<p>This is where multi agent systems stop being a buzzword and start being useful.</p>]]></description></item></channel></rss>