<?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>Forem: Christopher Nohall</title>
    <description>The latest articles on Forem by Christopher Nohall (@cnohall).</description>
    <link>https://forem.com/cnohall</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.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3799764%2Ffc476774-6ff8-4926-bbb2-f72319ed6200.png</url>
      <title>Forem: Christopher Nohall</title>
      <link>https://forem.com/cnohall</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://forem.com/feed/cnohall"/>
    <language>en</language>
    <item>
      <title>I built a prototype to classify API deprecations. Here's what I learned.</title>
      <dc:creator>Christopher Nohall</dc:creator>
      <pubDate>Sun, 01 Mar 2026 10:34:38 +0000</pubDate>
      <link>https://forem.com/cnohall/i-built-a-prototype-to-classify-api-deprecations-heres-what-i-learned-33pm</link>
      <guid>https://forem.com/cnohall/i-built-a-prototype-to-classify-api-deprecations-heres-what-i-learned-33pm</guid>
      <description>&lt;p&gt;I went looking for a tool that would tell me specifically when an API I depend on is being deprecated — not every changelog entry, just the ones that mean "you need to do something before this date or something breaks."&lt;/p&gt;

&lt;p&gt;It didn't exist in the way I wanted, so I built a proof of concept to see if the problem was solvable.&lt;/p&gt;

&lt;h2&gt;
  
  
  The problem with existing approaches
&lt;/h2&gt;

&lt;p&gt;The obvious workaround is subscribing to vendor mailing lists. The issue: when you depend on 10+ external APIs, you end up with 10 different newsletters. Breaking changes share an inbox with patch notes, SDK releases, and marketing fluff.&lt;/p&gt;

&lt;h2&gt;
  
  
  What the prototype does
&lt;/h2&gt;

&lt;p&gt;I wrote a script that fetches changelogs concurrently, then passes the content to an LLM to identify deprecations and ignore everything else.&lt;/p&gt;

&lt;p&gt;The classification works. "gpt-4-turbo is being deprecated" comes out labeled correctly. "We added an optional field" gets ignored. The hardest part is the parsing layer - Stripe uses structured JSON, but others use unstructured blog posts. I'm currently using a headless browser to normalize these into a consistent schema before classification.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I'm building
&lt;/h2&gt;

&lt;p&gt;Sunsetr acts as an "Uptime Monitor for the Future." It watches feeds for major providers and sends one clean alert when it detects a sunset.&lt;/p&gt;

&lt;p&gt;Landing page and waitlist at &lt;a href="https://sunsetr.nohall.dev" rel="noopener noreferrer"&gt;sunsetr.nohall.dev&lt;/a&gt;. Targeting $19/mo at launch.&lt;/p&gt;

&lt;p&gt;Is the RSS reader approach actually working for anyone, or is this a problem most teams just accept?&lt;/p&gt;

</description>
      <category>webdev</category>
      <category>api</category>
      <category>devtools</category>
      <category>showdev</category>
    </item>
  </channel>
</rss>
