Articles

Operator notes.

Problem-first writing on AI, automation, and the process underneath. Each piece names a real bottleneck, exposes the broken layer nobody owns, and shows what the working system actually looks like.

Enterprise AI

Why AI agents give inconsistent answers — and the semantic layer they're missing

A semantic layer for AI agents is the piece most enterprises skip: only 27% have a governed one, so agents reason confidently over data they misunderstand.

6 min read →
Commercial real estate

AI in commercial real estate: why 92% pilot and only 5% ship

AI in commercial real estate is everywhere in pilots, rarely in production. The blocker is your lease and maintenance data, not the model.

7 min read →
Logistics & supply chain

Agentic supply chain planning is only as good as the data your systems agree on

Agentic supply chain planning compresses replanning cycles. The operator read: it runs on siloed ERP/WMS/TMS data your own systems don't agree on.

7 min read →
AI governance

No-code AI agents are production integrations nobody owns

No-code AI agents like Amazon Quick make building one effortless. The operator read: an always-on agent is a production integration, and nobody owns what breaks.

6 min read →
Industrial ops

Industrial AI runs on an asset inventory most operators don't have

Accenture just paid ~$4.2B for OT asset visibility. The operator read: industrial AI fails on the asset inventory underneath it, not the model.

6 min read →
Manufacturing

AI quoting speeds up the manufacturing RFQ. The price is still only as good as your cost data.

AI quoting tools collapse the readable part of a manufacturing RFQ — reading the drawing, configuring the BOM. The part that decides whether you make money, they can't. The operator read.

6 min read →