§ Research

Long-form essays.

A permanent home for the essays I intend to hold up over time — closer to whitepapers than to blog posts.

2026
  1. An authority model for agent fleets: promotion by boring reliability, demotion at the speed of doubt, and a break-glass lever for 2am. Draft.

    #agents#governance#ax
  2. Permissions for agents that remember: classify at the write path, filter before fusion, fail secure, and never let credentials in the door. Draft.

    #security#agents#memory#permissions
  3. Organizational memory needs two independent axes — how much to trust a fact, and who may see it — with opposite enforcement mechanisms. Draft.

    #memory#governance#epistemics#agents
  4. On owning your tools, your keys, your data, and the systems that shape your work. Draft — being rewritten from the actual practice.

    #sovereignty#nostr#practice
  5. The environments agents work in matter more than the models working in them. Notes toward a discipline. Draft.

    #agents#ax#design
  6. Retrieval, memory and provenance are not features on top of a model. They are the load-bearing structure. Draft.

    #agents#context#infrastructure
  7. Retrieval solves a different problem than memory. I ended up running four services to keep them apart. Draft.

    #memory#rag#architecture#agents