From Vibe Coding to AI Operations
The seven principles I use to turn Claude Code, Codex and scheduled automation into a governed engineering workspace.
The AI Agent Architect
The model is 5% of an enterprise AI project. Identity, permissions, memory, audit, cost. The other 95% decides whether it ships, and it's where the careers are going. Practitioner written from inside real deployments. Start with two free field books.
Live Last read · last published · next check
Latest posts
The Model Is the Least Interesting Part of This AI Project
I’m helping a friend build a new system for his business with a Claude Code engineering team. Almost everything that matters sits outside the model.
The First Question Isn’t Which Model. It’s Which Country
A UAE tender assumed in-country AI inference was broadly available. The Jurisdiction Screen found a much narrower answer.
The AI Agent Tax Nobody Talks About
Agentic autonomy has a price when you're not sat at your desk.
Is Your AI Project Really an AI Project?
Read the verbs before anybody prices the architecture.
Your AI Agent Has No ROI. That's Why You Want a Kill Switch
Nobody caps a Facebook campaign that returns $1.50 on the dollar.
5 Signs Your AI Project Is Dead on Arrival
Each one is readable from the business case. None of them is the model you chose.
How I Really Use Claude Code: Operate, Don't Generate
Nine months of building with a recurring requirement has changed my relationship with Claude Code.
Can ChatGPT Replace Power BI? Someone's About to Ask You.
Every BI vendor has spent eighteen months rebuilding around conversational AI.
I Ran a Bank's Risk Model On An AI Agent
Just because it passes functional tests doesn't mean it's financially viable