Summary: AI is making me more productive while simultaneously making me less inclined to publish the results.
Summary: Agent orchestration doesn’t need another custom harness: a portable session manager, existing coding agents, and a thin skill can be enough.
Summary: How I added llms.txt and Markdown exports to my Pelican blog, giving LLMs and other tools a cleaner, machine-readable way to access its content.
Summary: An intentionally minimal AI evals runner that illustrates how to test LLMs.
Summary: Turning MicroGPT into a more educational, configurable project and finding that modifying and experimenting with it was the best way to understand how LLMs work.