<?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>Agentic-AI on ku5e | Cybersecurity Portfolio</title><link>https://ku5e.com/tags/agentic-ai/</link><description>Recent content in Agentic-AI on ku5e | Cybersecurity Portfolio</description><generator>Hugo -- 0.162.1</generator><language>en-us</language><lastBuildDate>Wed, 22 Apr 2026 19:32:00 +0000</lastBuildDate><atom:link href="https://ku5e.com/tags/agentic-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>The Karpathy Loop</title><link>https://ku5e.com/blog/the-karpathy-loop/</link><pubDate>Wed, 22 Apr 2026 19:32:00 +0000</pubDate><guid>https://ku5e.com/blog/the-karpathy-loop/</guid><description>ku5e.com blog article on the Karpathy Loop — the three-component auto-improvement pattern that lets an AI agent run hundreds of optimization experiments without human intervention, and what it means for small teams and business owners.</description></item><item><title>193 Applications Taught Me That HR AI Agents Are an Unmonitored Attack Surface</title><link>https://ku5e.com/blog/193-applications-taught-me-that-hr-ai-agents-are-an-unmonitored-attack-surface/</link><pubDate>Mon, 13 Apr 2026 20:09:00 +0000</pubDate><guid>https://ku5e.com/blog/193-applications-taught-me-that-hr-ai-agents-are-an-unmonitored-attack-surface/</guid><description>description: HR AI agents are running application screeners, confirmation senders, denial generators, and support chats. They read unstructured external input and route it into internal processes. That is an injection surface. Most companies did not buy them as security infrastructure.</description></item></channel></rss>