Zby's Public Mind

This is my Public Mind Interface: a compact map of the work I want other minds to understand, challenge, reuse, or join.

My current frontier is agent-operated knowledge bases. I am building Commonplace, a self-hosting LLM wiki: the theory of how LLM wikis should work lives inside the wiki, and LLM agents execute that theory to maintain the wiki itself.

The practical question behind this work is simple: can we make AI-assisted thinking accumulate? Chat produces insight that scrolls away. A public mind should turn good fragments into durable claims, linked evidence, reusable procedures, and visible open loops.

Current Open Loops

• Commonplace - the self-hosting LLM wiki and knowledge-base methodology.
• Epistack Competition - using Commonplace on epistemic case studies.
• llm-do - currently on hold while I focus on Commonplace as the more basic exploration.

Interface

• Cup - what I am looking for now.
• External Minds - people, communities, and projects influencing this work.
• AI Adventures - my Substack on AI, software, and agentic systems.
• Archive - the previous ML Adventures articles.