<?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>Cuda on Bits and Bytes</title>
    <link>https://blog.coder3101.com/tags/cuda/</link>
    <description>Recent content in Cuda on Bits and Bytes</description>
    <generator>Hugo -- 0.152.2</generator>
    <language>en</language>
    <lastBuildDate>Sun, 05 Apr 2026 00:00:00 +0500</lastBuildDate>
    <atom:link href="https://blog.coder3101.com/tags/cuda/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Understanding CUDA Kernels</title>
      <link>https://blog.coder3101.com/posts/understanding-cuda-kernels/</link>
      <pubDate>Sun, 05 Apr 2026 00:00:00 +0500</pubDate>
      <guid>https://blog.coder3101.com/posts/understanding-cuda-kernels/</guid>
      <description>&lt;p&gt;At work, we focus on optimizing LLM serving, and one topic that comes up repeatedly is kernel optimization. I want to share some insights into what a kernel actually is and where it fits in the stack, because believe it or not, every modern LLM and diffusion model is ultimately powered by kernels running on a GPU.&lt;/p&gt;
&lt;p&gt;I have some familiarity with Compute Unified Device Architecture (CUDA) and I also happen to have an NVIDIA Blackwell GPU in my workstation, so in this post I will explain what a kernel is and walk through writing one from scratch in CUDA.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
