<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Architecture on Technical Blog</title>
    <link>https://hugo-blog-923.pages.dev/tags/architecture/</link>
    <description>Recent content in Architecture on Technical Blog</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sat, 04 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://hugo-blog-923.pages.dev/tags/architecture/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Internal-docs RAG: chat ingress, vector search, and an async judge loop</title>
      <link>https://hugo-blog-923.pages.dev/posts/internal-docs-rag-architecture-notes/</link>
      <pubDate>Sat, 04 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://hugo-blog-923.pages.dev/posts/internal-docs-rag-architecture-notes/</guid>
      <description>&lt;p align=&#34;center&#34;&gt;
  &lt;img src=&#34;pexels_10376254.jpg&#34; alt=&#34;Overhead view of a professional at a desk with laptop, tablet, and papers — internal work and digital tools.&#34; style=&#34;max-width: min(100%, 820px); height: auto;&#34; /&gt;
&lt;/p&gt;
&lt;p&gt;This note describes an &lt;strong&gt;internal RAG&lt;/strong&gt; pattern for &lt;strong&gt;policy and handbook-style documents&lt;/strong&gt;: employees ask questions in a familiar chat surface, the backend retrieves by &lt;strong&gt;semantic similarity&lt;/strong&gt;, and a &lt;strong&gt;separate evaluation path&lt;/strong&gt; scores answers for quality and traceability. The layout maps cleanly to typical &lt;strong&gt;AWS&lt;/strong&gt; building blocks (API Gateway, Lambdas, object storage, a vector index, DynamoDB, and a queue).&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
