<?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>Systems on DBJ Iceberg</title><link>https://iceberg.dbj.org/tags/systems/</link><description>Recent content in Systems on DBJ Iceberg</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 10 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://iceberg.dbj.org/tags/systems/index.xml" rel="self" type="application/rss+xml"/><item><title>The Language Layer Does Not Matter</title><link>https://iceberg.dbj.org/posts/language-layer-does-not-matter/</link><pubDate>Tue, 10 Mar 2026 00:00:00 +0000</pubDate><guid>https://iceberg.dbj.org/posts/language-layer-does-not-matter/</guid><description>On llama.cpp, inference engines, and why the host language is irrelevant when the GPU is the bottleneck.</description></item></channel></rss>