Writing

Putting AI agents to real work.

Long-form on what it actually takes to hand operational work to AI agents, the systems, the guardrails, and the lessons, written from the seat of someone doing it every day.

The second-brain operating system

Helping My Wife Build a Playlist Builder With Her AI Assistant

She teaches spin and rebuilds her class playlist every month. Handing that job to her own AI assistant, what stays hers, and the Spotify endpoint that stopped publishing the one number it depends on.

September 8, 20269 min read
The second-brain operating system

Write Your AI Agent Doctrine, Not Prompts

A capable AI agent keeps getting your system subtly wrong because it reasons from scratch instead of reading it. The two-layer fix, doctrine and a primer, from running one agent across several businesses.

July 26, 20268 min read
Agentic commerce

Building a Product Feed for AI Shopping Agents

The buyer evaluating your catalog is becoming an AI agent, not a person. How I'm enriching a B2B product feed for that shift, and why machine-readable data is starting to decide the sale.

July 23, 20268 min read
Agents with real authority

What I Actually Let an AI Agent Do Without Me

The question isn't what an AI agent can do - it's what you let it do without you. The loop it runs, how I handed it over in stages, and the reversibility test that draws the line.

July 17, 20267 min read
Agents with real authority

What It Actually Costs to Run an AI Agent That Spends Real Money

I have an AI agent that runs an ad account on its own, with real budget and real bids. What it costs, how it earned the authority to spend without asking, and where a human still holds the line.

July 15, 20267 min read
The second-brain operating system

An AI Agent Runs My Workday. Here's How It Actually Works.

I handed daily operations to an AI agent I call Atlas: how it reconstructs the day from the tools I already use, what it decides on its own, and where a human still holds the line.

July 12, 20267 min read

More on the way.