<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Jonathan Mann — Blog</title><description>Notes on engineering leadership, SaaS architecture, and fractional CTO work.</description><link>https://jonathanmann.tech/</link><item><title>Keep Your Own Data: A Personal Data Sovereignty Stack for 2026</title><link>https://jonathanmann.tech/blog/keep-your-own-data/</link><guid isPermaLink="true">https://jonathanmann.tech/blog/keep-your-own-data/</guid><description>The asymmetry no one&apos;s fixing: every AI company has your data and you don&apos;t. Here&apos;s the personal data sovereignty stack I&apos;m building on an Intel Arc Pro B70 and Qwen3.6-35B-A3B.</description><pubDate>Sat, 23 May 2026 00:00:00 GMT</pubDate></item><item><title>Hello, again</title><link>https://jonathanmann.tech/blog/hello-world/</link><guid isPermaLink="true">https://jonathanmann.tech/blog/hello-world/</guid><description>Kicking off a writing habit on engineering leadership, fractional CTO work, and the strange shape of small technical teams.</description><pubDate>Thu, 21 May 2026 00:00:00 GMT</pubDate></item><item><title>Running Qwen3.6 Locally on the Intel Arc Pro B70: Why I Chose llama.cpp Over vLLM</title><link>https://jonathanmann.tech/blog/intel-arc-pro-b70-local-llm-llamacpp-qwen3/</link><guid isPermaLink="true">https://jonathanmann.tech/blog/intel-arc-pro-b70-local-llm-llamacpp-qwen3/</guid><description>Real benchmarks running Qwen3.6-35B-A3B on the Intel Arc Pro B70. Why llama.cpp beats vLLM for this model today, and what&apos;s coming next.</description><pubDate>Thu, 21 May 2026 00:00:00 GMT</pubDate></item></channel></rss>