<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Posts on DimonBlog</title><link>https://dimonb.com/categories/posts/</link><description>Recent content in Posts on DimonBlog</description><generator>Hugo</generator><language>en-us</language><managingEditor>blog@dimonb.com (Dmitrii Balabanov)</managingEditor><webMaster>blog@dimonb.com (Dmitrii Balabanov)</webMaster><lastBuildDate>Fri, 31 Jul 2026 19:10:00 +0300</lastBuildDate><atom:link href="https://dimonb.com/categories/posts/index.xml" rel="self" type="application/rss+xml"/><item><title>A Loop, Not a Pipeline: How to Design an Autonomous Full-Cycle Software Development System</title><link>https://dimonb.com/posts/a-loop-not-a-pipeline/</link><pubDate>Fri, 31 Jul 2026 19:10:00 +0300</pubDate><author>blog@dimonb.com (Dmitrii Balabanov)</author><guid>https://dimonb.com/posts/a-loop-not-a-pipeline/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Agents can write code, but autonomous software development is not a code-generation problem. It is a control-systems problem: designing feedback loops, independent verification, repair, and measured autonomy.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;Agents already write most of the code at frontier companies, yet no system in the world runs the path &amp;ldquo;idea → code → production → operations&amp;rdquo; without a human. This article argues that the bottleneck is not model intelligence but the engineering of feedback loops, and derives four design rules for an autonomous full-cycle system. The frame comes from reliability theory and cybernetics of the 1950s-70s — where &amp;ldquo;build reliable systems out of unreliable components&amp;rdquo; was solved once already — combined with the LLM-agent research of 2023-2026, where it is being solved a second time.&lt;/p&gt;</description></item></channel></rss>