<?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>Ptu on Ricardo Martins — Cloud Architecture, Azure, Kubernetes &amp; AI</title>
    <link>https://rmmartins.com/tags/ptu/</link>
    <description>Recent content in Ptu on Ricardo Martins — Cloud Architecture, Azure, Kubernetes &amp; AI</description>
    <image>
      <title>Ricardo Martins — Cloud Architecture, Azure, Kubernetes &amp; AI</title>
      <url>https://rmmartins.com/images/profile.png</url>
      <link>https://rmmartins.com/images/profile.png</link>
    </image>
    <generator>Hugo</generator>
    <language>en-US</language>
    <lastBuildDate>Wed, 06 May 2026 21:50:50 -0400</lastBuildDate>
    <atom:link href="https://rmmartins.com/tags/ptu/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Azure OpenAI in production: tokens, throughput, and high availability</title>
      <link>https://rmmartins.com/2026/06/19/azure-openai-in-production-tokens-throughput-and-high-availability/</link>
      <pubDate>Fri, 19 Jun 2026 10:00:00 -0400</pubDate>
      <guid>https://rmmartins.com/2026/06/19/azure-openai-in-production-tokens-throughput-and-high-availability/</guid>
      <description>HTTP 429 isn&amp;#39;t a bug, it&amp;#39;s bad capacity planning. Deployment types, PTU vs Standard, multi-region, retry patterns, and how not to take down your chatbot on launch day.</description>
    </item>
  </channel>
</rss>
