<?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>Product on Muhammad Hassan Raza</title><link>https://mhassan.dev/categories/product/</link><description>Recent content in Product on Muhammad Hassan Raza</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Fri, 25 Sep 2026 10:00:00 +0500</lastBuildDate><atom:link href="https://mhassan.dev/categories/product/index.xml" rel="self" type="application/rss+xml"/><item><title>LLMs Can't Save Bad UX</title><link>https://mhassan.dev/blog/llms-cant-save-bad-ux/</link><pubDate>Wed, 08 Apr 2026 10:00:00 +0500</pubDate><guid>https://mhassan.dev/blog/llms-cant-save-bad-ux/</guid><description>When an AI feature flops, the model is rarely the problem. Three features I watched fail, what fixed them, and the questions I now ask before anyone writes a prompt.</description></item></channel></rss>