<?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>Kernel-Fusion on martinuke0&#39;s Blog</title>
    <link>https://martinuke0.github.io/tags/kernel-fusion/</link>
    <description>Recent content in Kernel-Fusion on martinuke0&#39;s Blog</description>
    <image>
      <title>martinuke0&#39;s Blog</title>
      <url>https://martinuke0.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</url>
      <link>https://martinuke0.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</link>
    </image>
    <generator>Hugo -- 0.152.2</generator>
    <language>en</language>
    <lastBuildDate>Mon, 07 Sep 2026 10:00:29 +0000</lastBuildDate>
    <atom:link href="https://martinuke0.github.io/tags/kernel-fusion/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Scaling TensorRT-LLM: Kernel Fusion and Inflight Batching in Production</title>
      <link>https://martinuke0.github.io/posts/2026-09-07-scaling-tensorrt-llm-kernel-fusion-and-inflight-batching-in-production/</link>
      <pubDate>Mon, 07 Sep 2026 10:00:29 +0000</pubDate>
      <guid>https://martinuke0.github.io/posts/2026-09-07-scaling-tensorrt-llm-kernel-fusion-and-inflight-batching-in-production/</guid>
      <description>A practitioner&amp;#39;s deep dive into how TensorRT-LLM uses kernel fusion and inflight batching to push GPU utilization past 80% in production LLM serving.</description>
    </item>
  </channel>
</rss>
