<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<title>sysrex - AI</title>
	<subtitle>This is mostly about the things that I am learning and don&#39;t want to forget just yet!</subtitle>
	<link href="https://sysrex.com/tags/ai/atom.xml" rel="self" type="application/atom+xml"/>
	<link href="https://sysrex.com"/>
	<generator uri="https://www.getzola.org/">Zola</generator>
	<updated>2025-07-22T00:00:00+00:00</updated>
	<id>https://sysrex.com/tags/ai/atom.xml</id>
	<entry xml:lang="en">
		<title>What LLM Evals Actually Are (and Why &#39;Looks Good to Me&#39; Isn&#39;t a Strategy)</title>
		<published>2025-07-22T00:00:00+00:00</published>
		<updated>2025-07-22T00:00:00+00:00</updated>
		<link href="https://sysrex.com/posts/what-are-llm-evals/" type="text/html"/>
		<id>https://sysrex.com/posts/what-are-llm-evals/</id>
			
		
		<summary>A lot of teams building on top of LLMs ship a prompt, look at a handful of
outputs, decide it &quot;seems…</summary>
		
	</entry>
	<entry xml:lang="en">
		<title>LLM Quantization, Explained in Plain English</title>
		<published>2025-03-12T00:00:00+00:00</published>
		<updated>2025-03-12T00:00:00+00:00</updated>
		<link href="https://sysrex.com/posts/llm-quantization-explained/" type="text/html"/>
		<id>https://sysrex.com/posts/llm-quantization-explained/</id>
			
		
		<summary>If you&apos;ve tried running a large language model on your own hardware, you&apos;ve
probably run into the sa…</summary>
		
	</entry>
</feed>
