Welcome to the Lyceum
Aristotle taught while walking. The Lyceum wasn’t a lecture hall — it was a covered walkway where questions were pursued in motion, back and forth, with the student doing as much work as the teacher.
That is the model for this site.
Lyceum Agents is a school for the age of agents: short essays on how agentic AI systems actually work, how to build them, and — inseparably — how to build them responsibly. The two subjects are printed on the same page here on purpose. An architecture decision is an ethics decision the moment software starts taking actions instead of just producing text.
how this place works
A few deliberate choices, so you know what you’re standing on:
- Light on media, heavy on interaction. No hero videos, no stock illustrations. Where another site would embed a diagram, this one will hand you a dialogue to steer or a terminal to type into.
- The shell is the front door. Press
/anywhere and typehelp. You canlsthe posts,openanything by slug, and discover a few things that aren’t on any menu. - Everything compounds. Posts are numbered days in a longer curriculum. Over time they will be woven into a virtual textbook — the essays are the first draft of its chapters.
Here is what that looks like in practice — a real turn, machinery included:
what’s coming
Three threads, braided:
- Foundations — what agents are, planning loops, tool use, memory, multi-agent orchestration.
- Ethics — delegation and accountability, alignment in practice, what we owe the people on the other side of an agent’s actions.
- Tools — open-source instruments built alongside the writing, starting with melchizedek, a multi-model multi-agent orchestration framework.
Start with What is an agent? — it ends with a conversation rather than a conclusion.