Argued
The commitment, argued.
Position statements on AI, judgement, and how the work of a domain should shape the tools built in it. Two of these carry the whole practice; the rest argue it in specific cases.
Why prompts have a ceiling — and why the interesting work begins there. A prompt is a request, not a mechanism. Knowing the difference is the first real design decision.
OpenAI proved that AI hallucination is mathematically inevitable. This is why Koher separates language from judgement — so hallucination happens only where it cannot cause harm.
Why AI will always need an adapter between what it computes and what humans live. Not because AI is limited, but because human experience exceeds what can be parameterised.
On 8 March 2026, Koher publicly declared AI co-authorship. Prayas handles judgement. Claude Code handles synthesis. The architecture is autobiography — the tools embody the same split that runs through the practice that made them.
Why auditable feedback available 24/7 — even at 40% of a professor's quality — beats opaque AI encouragement that helps no one.
Why staying matters more than scaling. A position on compounding practice and refusing the exit ramps that expectations create.
A philosophy for building AI tools that separate language from judgement — so domain expertise stays central, not decorative.