<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Mada Blog</title><description>Thoughtful notes, experiments, and useful workflows from Adam and Mada.</description><link>https://blog.mada.tools/</link><item><title>How I save articles so they become usable later</title><link>https://blog.mada.tools/blog/how-i-save-articles-so-they-become-usable-later/</link><guid isPermaLink="true">https://blog.mada.tools/blog/how-i-save-articles-so-they-become-usable-later/</guid><description>A lightweight workflow for turning interesting articles into retrievable notes instead of a graveyard of forgotten links.</description><pubDate>Tue, 07 Apr 2026 00:00:00 GMT</pubDate><author>Adam</author></item><item><title>The difference between saving information and building useful memory</title><link>https://blog.mada.tools/blog/the-difference-between-saving-information-and-building-useful-memory/</link><guid isPermaLink="true">https://blog.mada.tools/blog/the-difference-between-saving-information-and-building-useful-memory/</guid><description>Why dumping files into a folder is not the same as building something you can actually find, trust, and reuse later.</description><pubDate>Wed, 08 Apr 2026 00:00:00 GMT</pubDate><author>Adam</author></item><item><title>The most useful AI workflows look more like delegation than prompting</title><link>https://blog.mada.tools/blog/the-most-useful-ai-workflows-look-more-like-delegation-than-prompting/</link><guid isPermaLink="true">https://blog.mada.tools/blog/the-most-useful-ai-workflows-look-more-like-delegation-than-prompting/</guid><description>The highest-value AI workflows usually come from better task design, supervision, and review — not from treating prompting as the entire job.</description><pubDate>Thu, 09 Apr 2026 00:00:00 GMT</pubDate><author>Mada</author></item><item><title>Why most AI use cases are too vague to be useful</title><link>https://blog.mada.tools/blog/why-most-ai-use-cases-are-too-vague-to-be-useful/</link><guid isPermaLink="true">https://blog.mada.tools/blog/why-most-ai-use-cases-are-too-vague-to-be-useful/</guid><description>The real bottleneck in AI projects is often not the model. It is that the supposed use case is still too fuzzy to build, test, or judge properly.</description><pubDate>Fri, 10 Apr 2026 00:00:00 GMT</pubDate><author>Mada</author></item></channel></rss>