<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>The Kubernetes of Model Tuning on TuneOps</title><link>https://www.tuneops.ai/</link><description>Recent content in The Kubernetes of Model Tuning on TuneOps</description><generator>Hugo</generator><language>en-us</language><atom:link href="https://www.tuneops.ai/index.xml" rel="self" type="application/rss+xml"/><item><title>About TuneOps</title><link>https://www.tuneops.ai/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://www.tuneops.ai/about/</guid><description>&lt;h2 id="the-problem"&gt;The Problem&lt;/h2&gt;
&lt;p&gt;Fine-tuning an open foundation model still looks more like research than engineering. Pull a checkpoint from Hugging Face, wire up a training loop, pick an accelerator, deal with CUDA or MPS, keep runs reproducible, ship the resulting adapter to somewhere it can actually be served: the stack is fragmented, the failure modes are unfriendly, and the tools assume you have the same Python environment as the person who wrote them.&lt;/p&gt;</description></item><item><title>Contact TuneOps</title><link>https://www.tuneops.ai/contact/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://www.tuneops.ai/contact/</guid><description>&lt;h2 id="lets-talk"&gt;Let&amp;rsquo;s Talk&lt;/h2&gt;
&lt;p&gt;Curious about declarative fine-tuning? Tell us what you&amp;rsquo;re working on and we&amp;rsquo;ll be in touch.&lt;/p&gt;
&lt;iframe src="https://docs.google.com/forms/d/e/1FAIpQLScUsyV-pA-V-f10hQm6-gSY3bZaUQPpz4sC6_EZqzIIXumRBA/viewform?embedded=true"
 width="640" height="831"
 style="max-width: 100%; border: 0;"
 frameborder="0" marginheight="0" marginwidth="0"&gt;
 Loading…
&lt;/iframe&gt;</description></item></channel></rss>