Summary: Agent orchestration doesn’t need another custom harness: a portable session manager, existing coding agents, and a thin skill can be enough.
Summary: An intentionally minimal AI evals runner that illustrates how to test LLMs.
Summary: A Javascript/NodeJS savegame editor for the Vampire Crawlers videogame.
Summary: My current unified agentic workflow, where Claude Code, Copilot, and local models act as interchangeable executors, driven by a shared configuration and centered around VS Code as the review surface.
Summary: Two recent book acquisitions: one about 80X86 Assembly Language, and the other about creating CLI tools in Go.