<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Homelab AI on Stanislav Cherkasov | {DevOps,DevSecOps,Platform}</title><link>https://tenhishadow.github.io/series/homelab-ai/</link><description>Recent content in Homelab AI on Stanislav Cherkasov | {DevOps,DevSecOps,Platform}</description><generator>Hugo -- gohugo.io</generator><language>en</language><managingEditor>svcdevops@tenhi.dev (Stanislav Cherkasov)</managingEditor><webMaster>svcdevops@tenhi.dev (Stanislav Cherkasov)</webMaster><copyright>© 2026 Stanislav Cherkasov</copyright><atom:link href="https://tenhishadow.github.io/series/homelab-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>One GPU, One LLM Lane: Operating an RTX PRO 4000 SFF in k3s</title><link>https://tenhishadow.github.io/posts/ai/2026-07-02-rtx-pro-4000-day-2-k3s-gpu-lane/</link><pubDate>Thu, 02 Jul 2026 00:00:00 +0000</pubDate><author>svcdevops@tenhi.dev (Stanislav Cherkasov)</author><guid>https://tenhishadow.github.io/posts/ai/2026-07-02-rtx-pro-4000-day-2-k3s-gpu-lane/</guid><description>A practical operating model for a single 24 GB GPU in k3s: one active LLM workload, explicit scheduling rules, preflight-gated model switches, and measured Qwen3.6 performance</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://tenhishadow.github.io/posts/ai/2026-07-02-rtx-pro-4000-day-2-k3s-gpu-lane/featured.png"/></item></channel></rss>