<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Alex Zerntev</title><description>Notes on building with coding agents, production agent systems, software architecture, and engineering judgment.</description><link>https://alexzerntev.github.io/</link><item><title>Reality Doesn’t Fit in a Prompt</title><link>https://alexzerntev.github.io/blog/reality-doesnt-fit-in-a-prompt/</link><guid isPermaLink="true">https://alexzerntev.github.io/blog/reality-doesnt-fit-in-a-prompt/</guid><description>Models can process a description of what is, but they do not share the lived context from which we decide what should be.</description><pubDate>Tue, 28 Jul 2026 00:00:00 GMT</pubDate><category>LLMs</category><category>AI agents</category><category>Human judgment</category></item><item><title>Rebuilding this site with coding agents</title><link>https://alexzerntev.github.io/blog/rebuilding-this-site-with-coding-agents/</link><guid isPermaLink="true">https://alexzerntev.github.io/blog/rebuilding-this-site-with-coding-agents/</guid><description>What stayed human while I rebuilt my website with coding agents - taste, vetoes, and a decision log that kept rejected directions from coming back.</description><pubDate>Tue, 21 Jul 2026 00:00:00 GMT</pubDate><category>Coding agents</category><category>Engineering judgment</category></item></channel></rss>