Insights.

Notes on the ideas behind Manwe. Ontology, sovereignty and research, and how the harness puts them to work.

Product

What it means for an agent to stay

Most AI "agents" are a prompt that disappears when the chat closes. A persistent advisor keeps its role, its lessons and its watches, and comes back when something changes.

31 October 2026 · 2 min read
Essay

Your AI doesn't need a longer memory. It needs an ontology.

Chat history is a transcript, not an understanding. What makes context useful is knowing what things are, how they relate and why they matter.

30 October 2026 · 3 min read
Guide

Worlds and Moons: context for a life with many projects

Your company, your studies and your home renovation shouldn't share one memory. A practical guide to giving each part of your life its own context, and knowing when to split it further.

28 October 2026 · 2 min read
Research

Three models, one question: why disagreement is the useful part

A single confident answer hides its weak points. Putting the same question to several models shows you where the reasoning is solid and where it's a guess.

27 October 2026 · 2 min read
Essay

AI sovereignty starts on the computer you own

Sovereignty isn't a slogan about where data centres are. For a person or a small team, it comes down to three practical questions about your context, your models and your exits.

24 October 2026 · 3 min read
Guide

How to hand an AI your research materials

The quality of AI research depends less on the model than on what you give it and what you ask for. A practical guide to briefs, sources, folders and checking the result.

20 October 2026 · 2 min read