<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Ai-Agent on nbdawn's Blog</title><link>https://blog.nbdawn.com/tags/ai-agent/</link><description>Recent content in Ai-Agent on nbdawn's Blog</description><generator>Hugo -- 0.163.3</generator><language>en</language><copyright>DJ.Kim 2025</copyright><lastBuildDate>Tue, 23 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.nbdawn.com/tags/ai-agent/index.xml" rel="self" type="application/rss+xml"/><item><title>Hunting the Repetition Loop in a Self-Hosted LLM Agent</title><link>https://blog.nbdawn.com/posts/hunting-the-repetition-loop-in-a-self-hosted-llm-agent/</link><pubDate>Tue, 23 Jun 2026 00:00:00 +0000</pubDate><guid>https://blog.nbdawn.com/posts/hunting-the-repetition-loop-in-a-self-hosted-llm-agent/</guid><description>&lt;h2 id="when-the-agent-kept-repeating-itself"&gt;When the Agent Kept Repeating Itself&lt;/h2&gt;
&lt;p&gt;At first I thought a request had hung. Where a tool call should have been, the model was instead generating its way toward &lt;code&gt;max_tokens&lt;/code&gt; and getting nowhere — sometimes repeating the same sentence over and over, other times just producing low-value filler that never resolved into the JSON the tool call needed. Either way it would burn through the token budget, occasionally time out, and take the whole agent loop down with it.&lt;/p&gt;</description></item></channel></rss>