<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>SDLC on Dimitri Shiliaev</title>
    <link>https://shiliaev.com/tags/sdlc/</link>
    <description>Recent content in SDLC on Dimitri Shiliaev</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Tue, 17 Mar 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://shiliaev.com/tags/sdlc/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Why 10x AI Code is Breaking Your Pipeline (And How to Fix It)</title>
      <link>https://shiliaev.com/blog/aicoding/</link>
      <pubDate>Tue, 17 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://shiliaev.com/blog/aicoding/</guid>
      <description>&lt;p&gt;If you spend any time on tech platforms or developer forums, you have undoubtedly heard the triumphant declaration: &lt;em&gt;Coding is solved.&lt;/em&gt;&lt;/p&gt;&#xA;&lt;p&gt;With the rise of autonomous agents and Large Language Models (LLMs), generating hundreds of lines of complex boilerplate now takes seconds. But here is the candid truth that engineering teams are learning the hard way: &lt;strong&gt;&amp;ldquo;Coding is solved&amp;rdquo; does not mean software engineering is solved.&lt;/strong&gt; Software engineering was never just about typing syntax. It is about architecture, integration, rigorous testing, and delivering a cohesive product. LLMs are probabilistic generators — they produce massive amounts of text with inherent uncertainty. You are no longer just giving rigid instructions; you are constantly probing a model, evaluating probabilistic results, and trying to align its output with your project&amp;rsquo;s strict reality.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
