Do I need to be technical?
No. You describe the work the way you would explain it to a new hire. I design the structure that runs it and hand you a guide written in plain language.
Will it do things without asking me?
No. Anything that spends money, goes to a customer, or cannot be undone waits for your yes. You set those boundaries before I build, and the gate is built into the structure rather than left to anyone's memory.
Can it build things, like a website or an app?
It can, if you direct it to. It runs on a coding agent, Claude Code, so it can take on a real build instead of only writing about one. A website or internal app is scoped separately from the department packages, so ask first. It plans first and waits for your approval.
What will actually change in my week?
Routine work finishes without asking you anything. Work is checked by a team lead inside the system before it reaches you, rather than by you at 11pm after it has already gone out. You read a short summary of what happened instead of a wall of chat history.
What do you need from me to start?
Answers to three questions: the process that eats your week, what must never go out without your approval, and which AI tools you already pay for. Ten to fifteen minutes.
How many agents will I actually have?
One Department has up to 4 agents and 1 team lead reviewer. Three Departments has up to 12 agents and 3 team lead reviewers. Full Organization has up to 5 departments, up to 25 agents, up to 5 team lead reviewers and a final adversary. Each team lead reviews its agents' output before it moves up.
Who owns what you build?
You do. Setup files sit on your own Claude Code or ChatGPT account and stay yours whether or not you keep working with me. I host nothing and license nothing back to you.
What happens to my data?
Your build is set up inside your own Claude Code or ChatGPT account. I work in it only while I build, and my access ends at handover. I don't take sensitive records such as patient information: I build and test with made-up examples, so your real data stays with you.
How much does it cost?
The consult is free. For a build you get a fixed price, in writing, by email before any call: tell me the work and I send the price and the delivery date, within one business day. The figure I send is the figure you pay for the scope we agree, and it does not change after you sign.
What happens after I get the price?
If it works for you, you pay it in full up front. The 5, 7 or 10 delivery days start on the day both your payment and your answers to the three intake questions are in. At handover you get the build and its four artifacts.
What does it cost to run, on top of what I pay you?
Model usage is billed by Anthropic or OpenAI directly, on your account, at their price. I add no markup and never handle it.
Is there anything to keep paying?
No. There is no retainer, no subscription and no license. You pay once for the build, the files are yours, and nothing switches off. For 14 days after handover I fix, at no charge, anything that does not work the way we agreed; after that, a fix or a change is quoted like new work.
Do you help after handover?
Yes. For 14 days after handover I fix, at no charge, anything that does not work the way we agreed. After that, a fix or a change is quoted like new work, and there is still nothing to pay monthly.
What happens when something breaks?
That's what the failure plan is for. It names what happens, who picks it back up, and how the department is restarted. It is one of the four artifacts you get with every build.
Still deciding?
Tell me the process. I'll say whether an organization is the right shape for it.
Get in touch