<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Henry Glendening</title><description>Software engineer building things at the intersection of mobile and AI.</description><link>https://henryg.dev/</link><language>en</language><item><title>Separating Surface Variation from Conceptual Variation in LLM Outputs</title><link>https://henryg.dev/posts/2026/human-touch-convergence/</link><guid isPermaLink="true">https://henryg.dev/posts/2026/human-touch-convergence/</guid><description>Across 3,300 LLM outputs, I measured three distinct layers of variation. Models rephrase freely while recycling the same concepts and arguments. Temperature doesn&apos;t help. A few structural prompt techniques do.</description><pubDate>Sun, 08 Mar 2026 12:00:00 GMT</pubDate><category>ai</category><category>creativity</category><category>llm</category><category>research</category><category>convergence</category><author>Henry</author></item><item><title>Persona Seeding, Memory Implants, and the Limits of LLM Creativity</title><link>https://henryg.dev/posts/2026/human-touch/</link><guid isPermaLink="true">https://henryg.dev/posts/2026/human-touch/</guid><description>I built a framework that places LLM agents in constrained creative environments with synthetic backstories and emotional memories, then measured whether it actually makes them more creative.</description><pubDate>Thu, 26 Feb 2026 00:00:00 GMT</pubDate><category>ai</category><category>creativity</category><category>llm</category><category>research</category><category>multi-agent</category><author>Henry</author></item><item><title>Hello, World</title><link>https://henryg.dev/posts/2026/hello-world/</link><guid isPermaLink="true">https://henryg.dev/posts/2026/hello-world/</guid><description>First post — what this site is about and what&apos;s coming.</description><pubDate>Mon, 16 Feb 2026 00:00:00 GMT</pubDate><category>intro</category><author>Henry</author></item></channel></rss>