<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Keller AI — Research Notes</title><description>DBA research on AI enterprise architecture by Daniel Keller. Frameworks, findings, and research translations for business leaders.</description><link>https://keller-ai.com/</link><language>en-gb</language><item><title>Hermes is Quietly Becoming the Agent to Watch</title><link>https://keller-ai.com/research/hermes-the-agent/</link><guid isPermaLink="true">https://keller-ai.com/research/hermes-the-agent/</guid><description>While everyone debates which frontier model to run, a multi-agent orchestration framework called Hermes has been maturing in the background. I have been running it. Here is what I think.</description><pubDate>Thu, 22 May 2025 00:00:00 GMT</pubDate></item><item><title>Prompt Engineering is Dead. Long Live Context Engineering.</title><link>https://keller-ai.com/research/prompt-engineering-is-dead/</link><guid isPermaLink="true">https://keller-ai.com/research/prompt-engineering-is-dead/</guid><description>The craft of writing clever prompts is being automated away. What actually determines output quality now is the context you give the model — and how you structure reusable work.</description><pubDate>Tue, 20 May 2025 00:00:00 GMT</pubDate></item><item><title>HTML vs Markdown: The Debate is Settled</title><link>https://keller-ai.com/research/html-vs-markdown/</link><guid isPermaLink="true">https://keller-ai.com/research/html-vs-markdown/</guid><description>A debate that ran hot through AI developer circles this year has reached a practical consensus. The answer is not one or the other — it is about where each belongs in the stack.</description><pubDate>Thu, 15 May 2025 00:00:00 GMT</pubDate></item></channel></rss>