<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Systems Thinking on Jamal Yusuf</title><link>https://jamal.dev/categories/systems-thinking/</link><description>Recent content in Systems Thinking on Jamal Yusuf</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 15 Jan 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://jamal.dev/categories/systems-thinking/index.xml" rel="self" type="application/rss+xml"/><item><title>Why Most Enterprise AI Agents Fail</title><link>https://jamal.dev/writing/why-enterprise-ai-agents-fail/</link><pubDate>Thu, 15 Jan 2026 00:00:00 +0000</pubDate><guid>https://jamal.dev/writing/why-enterprise-ai-agents-fail/</guid><description>&lt;p&gt;I have watched smart teams ship impressive agent demos — and then watch those same agents fail the only test that matters: &lt;strong&gt;would an expert trust this under pressure?&lt;/strong&gt;&lt;/p&gt;
&lt;p&gt;The model is rarely the villain. The retrieval design is.&lt;/p&gt;
&lt;h2 id="the-expert-retrieval-gap"&gt;The expert retrieval gap&lt;/h2&gt;
&lt;p&gt;Most enterprise AI agents are built around a comforting pipeline: chunk, embed, retrieve, generate. It looks scientific. It scales on slides. It also assumes experts think in paragraphs — flat, interchangeable, equally worthy of attention.&lt;/p&gt;</description></item></channel></rss>