<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agentic-Coding on Alex Herrero</title><link>https://alexherrero.dev/tags/agentic-coding/</link><description>Recent content in Agentic-Coding on Alex Herrero</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Fri, 01 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://alexherrero.dev/tags/agentic-coding/index.xml" rel="self" type="application/rss+xml"/><item><title>A 1M context window isn't free</title><link>https://alexherrero.dev/thoughts/1m-context-window-isnt-free/</link><pubDate>Fri, 01 May 2026 00:00:00 +0000</pubDate><guid>https://alexherrero.dev/thoughts/1m-context-window-isnt-free/</guid><description>&lt;p&gt;So agent harnesses like Claude and Antigravity recently introduced 1M-token context windows, which was great, at first. It turns out that a context window is re-sent on every turn, so a near-full window burns through your usage limit — FAST. Bummer.&lt;/p&gt;</description></item><item><title>Intro -&gt; Next Level</title><link>https://alexherrero.dev/thoughts/up-another-level/</link><pubDate>Sat, 25 Apr 2026 00:00:00 +0000</pubDate><guid>https://alexherrero.dev/thoughts/up-another-level/</guid><description>Nearly two years ago, early agentic coding let us build an on-device agent — in just a week — that would have taken months. I knew then that software was moving to the next level — today we call it agentic coding. Here&amp;rsquo;s where I&amp;rsquo;ll write about what that&amp;rsquo;s actually like.</description></item></channel></rss>