<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Hrithic Rajpurohit</title>
    <description>The latest articles on DEV Community by Hrithic Rajpurohit (@hrithic_boit).</description>
    <link>https://dev.to/hrithic_boit</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4230085%2F7204eb07-5177-49f2-9c8a-89031585da81.png</url>
      <title>DEV Community: Hrithic Rajpurohit</title>
      <link>https://dev.to/hrithic_boit</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/hrithic_boit"/>
    <language>en</language>
    <item>
      <title>The model that writes a page shouldn't be the one that signs it off</title>
      <dc:creator>Hrithic Rajpurohit</dc:creator>
      <pubDate>Thu, 12 Nov 2026 12:04:51 +0000</pubDate>
      <link>https://dev.to/hrithic_boit/the-model-that-writes-a-page-shouldnt-be-the-one-that-signs-it-off-48ol</link>
      <guid>https://dev.to/hrithic_boit/the-model-that-writes-a-page-shouldnt-be-the-one-that-signs-it-off-48ol</guid>
      <description>&lt;p&gt;My name is Hrithic Rajpurohit, and I work on the engineering team at Produsa Labs. We built Produsa, an AI website builder made for India, so I should tell you that before I describe how the build pipeline works.&lt;/p&gt;

&lt;p&gt;A website build is not one task. Research, brand, copy, layout, code and a final check all ask different things of a model. If the whole job goes through a single model, the writing, the design sense and the engineering quality all settle at whatever that model is average at.&lt;/p&gt;

&lt;p&gt;Produsa instead sends each job to the model that does it best. The ones we use are Claude, OpenAI, Gemini, xAI/Grok and Kimi. We have not published a map of which model takes which step, so I will not invent one here. The useful part is the idea: pick the model for the job, not one model for the whole product.&lt;/p&gt;

&lt;p&gt;We also added a second, separate pass. After a page is written and designed, a different AI reviews it against a written design rulebook. The first pass can look finished and still miss the brief, repeat a layout, or drift from the brand. That review only works if the rulebook is specific enough to fail. A vague "make it look good" rule just gets a yes.&lt;/p&gt;

&lt;p&gt;What I would tell another team doing this: write the bar down before you hook up the reviewer. If the only standard lives in someone's head, the review pass has nothing firm to argue with.&lt;/p&gt;

&lt;p&gt;If you have tried a reviewer model on generated UI or copy, I am curious what you put in the rulebook, and what it still lets through.&lt;/p&gt;

&lt;p&gt;This article was drafted with AI assistance and then checked by me. I can stand behind the facts above.&lt;/p&gt;

</description>
      <category>architecture</category>
      <category>ai</category>
      <category>webdev</category>
      <category>llm</category>
    </item>
  </channel>
</rss>
