<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Code-Review on David Lang</title>
    <link>https://www.davidlang.tech/tags/code-review/</link>
    <description>Recent content in Code-Review on David Lang</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sun, 22 Sep 2024 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://www.davidlang.tech/tags/code-review/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>AI-Powered Code Review: Integrating LLMs into Dev Workflows</title>
      <link>https://www.davidlang.tech/posts/ai-powered-code-review-integrating-llms-into-dev-workflows/</link>
      <pubDate>Sun, 22 Sep 2024 00:00:00 +0000</pubDate>
      <guid>https://www.davidlang.tech/posts/ai-powered-code-review-integrating-llms-into-dev-workflows/</guid>
      <description>&lt;p&gt;LLMs can summarize diffs, flag security smells, and suggest tests-but they should augment human review, not replace it.&lt;/p&gt;&#xA;&lt;h2 id=&#34;ci-integration&#34;&gt;CI Integration&lt;/h2&gt;&#xA;&lt;p&gt;Post PR diffs to an LLM with a structured prompt. Output JSON findings consumed by GitHub Actions or GitLab CI. Fail builds only on high-severity, high-confidence issues to reduce noise.&lt;/p&gt;&#xA;&lt;h2 id=&#34;prompt-design-for-reviews&#34;&gt;Prompt Design for Reviews&lt;/h2&gt;&#xA;&lt;p&gt;Include: changed files, diff hunks, coding standards doc, and explicit instruction to cite line numbers and avoid nits.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
