<?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>Infrastructure on Chirag Hasija</title>
    <link>https://chiraghasija.cc/tags/infrastructure/</link>
    <description>Recent content in Infrastructure on Chirag Hasija</description>
    <image>
      <title>Chirag Hasija</title>
      <url>https://chiraghasija.cc/og-image.png</url>
      <link>https://chiraghasija.cc/og-image.png</link>
    </image>
    <generator>Hugo -- 0.155.3</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 30 Oct 2025 08:47:00 +0530</lastBuildDate>
    <atom:link href="https://chiraghasija.cc/tags/infrastructure/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Terraform vs Pulumi in 2026: The IaC Showdown Is Over</title>
      <link>https://chiraghasija.cc/posts/terraform-vs-pulumi-iac-showdown-2026/</link>
      <pubDate>Thu, 30 Oct 2025 08:47:00 +0530</pubDate>
      <guid>https://chiraghasija.cc/posts/terraform-vs-pulumi-iac-showdown-2026/</guid>
      <description>Terraform, OpenTofu, and Pulumi in 2026: which IaC tool to use after the Terraform license change and the emergence of OpenTofu. A practical guide to making the right choice.</description>
    </item>
    <item>
      <title>How Meta Trains LLaMA 4 on 100,000 GPUs</title>
      <link>https://chiraghasija.cc/posts/how-meta-trains-llama4-100k-gpus/</link>
      <pubDate>Sat, 09 Aug 2025 09:42:00 +0530</pubDate>
      <guid>https://chiraghasija.cc/posts/how-meta-trains-llama4-100k-gpus/</guid>
      <description>An exploration of the distributed systems and hardware engineering required to train LLaMA 4 on tens of thousands of GPUs, and what limits performance at this scale.</description>
    </item>
  </channel>
</rss>
