Space frame canopy drawing with prairie-school building
Services

Five ways in. Every one ends with a system your team runs.

Every engagement is fixed scope, priced before any code, and tracked in a private client portal where you see the plan, the progress and what I need from you every week.

Built for companies of 10 to 500 people that run on software they did not write and are now adding AI to it. Larger companies come by referral.

Review a system someone built with AI

For a vibe-coded app or an AI tool the business already runs on.

What I do

I read every line and check it against the ten things a full software team would have covered: how your data is stored and who can reach it, sign-in and roles, input validation, exposed keys, compliance, notification triggers, workflows, reporting, logging, and documentation.

What you get

A plain-English report that rates each item safe, fix soon or fix now, a ranked fix list, and the critical fixes done. Your team keeps it safe from there.

How it starts

Read-only access to the code and hosting. The report comes back within two weeks.

Starts at $3,000. Most reviews run two weeks. Every engagement is a fixed price, built on $150 an hour, agreed before any work starts.

Start here
A boxy AI agent waves one hand, with handwritten notes labeling its head, joints and legs.

Oversight for staff who build with AI

For companies where employees already build tools with ChatGPT, Claude or Copilot.

What I do

I find every tool your staff built and what data it touches, set simple rules for what they may build and where, review each existing tool, and train your builders on the parts nobody told them were there.

What you get

An inventory, a one-page set of rules, each tool reviewed, and a standing monthly review reported in your portal. Your people keep building. What they build is safe to run the business on.

How it starts

A one-hour conversation with the people doing the building.

Start here

Build a new system or AI agents

For a part of the business that eats hours every week, or AI you want inside your own product.

What I do

I write down what the system has to do and how we will prove it works before any code. Then AI plans and writes, and I review every change and test it on your real cases. You watch it happen in your portal.

What you get

A working system on your real data, documented, with runbooks and working sessions so your team owns it. The success measure is set in Discovery: hours saved, cost saved or revenue added.

How it starts

Discovery, one conversation, no charge.

Start here

Connect AI to the systems you already have

For data trapped in older software, spreadsheets, or a system that should be replaced.

What I do

I map what data exists and who owns it, get it out cleanly on a schedule, and put it in one place your new tools can read. Where connecting would cost more than replacing, I say so and quote both.

What you get

The manual re-entry gone, automated workflows running on their own, and a runbook your team can follow.

How it starts

A list of the systems involved and one example of the work you do by hand today.

Start here

Be in the answer when customers ask AI who to call

For businesses whose customers ask ChatGPT, Claude or Perplexity for a recommendation.

What I do

I record what the AI models say about you today, make your site readable and quotable by them, build a tool that keeps it current, and ask the same questions again.

What you get

A before-and-after report showing how often you appear in AI answers for the questions your customers ask.

How it starts

Ten questions your customers ask when they are looking for a business like yours.

Start here

Not sure which one?

Start with Discovery. One conversation, no charge. I will tell you which of the five fits, or whether AI is the wrong tool for the problem.

Tell me what the system needs to do