Projects
These case studies show how I work and some of the projects I've worked on.
PT Max is a fitness-coaching platform for personal trainers and their clients, delivered through Nhance Group. I'm the architect and technical lead in the team behind it. Much of it was built with a team of AI coding agents (Anthropic's Claude Code) working in parallel, and nothing reaches production without a person signing it off.
It runs on the web at ptmax.app and on Android through Google Play, where it has been on the production track since August 2026. It takes card payments through Stripe, and real coaches and their clients use it.
The platform is built on .NET 10. A Blazor web app (server and WebAssembly) and a .NET MAUI Android app share one codebase, with SQL Server behind them, Stripe and Google Play Billing for payments, and Cloudflare and IIS in front.
There have been 6,394 commits since the first one on 2 February 2026, and Claude wrote 5,205 of them (81%), each co-authored by the agent. Ten or more agents usually work at once, each with its own job, branch and working copy, and as many as 13 have worked on it in a single day.
The code is covered by 9,456 automated .NET tests and 1,381 browser tests. The full browser suite runs before every release, and it scales to use every core of a modern computer, with successful runs on up to 32 in parallel.
The harder part of working with AI agents is trusting them with a live product that people pay for. I designed the rules and safeguards that make that possible.
One coordinating agent shares out the development database, ports, test phones and the production server, and keeps a record of every decision. Each working agent owns a branch, merges its own work and reports back. Anything that can't be undone needs a person's direct approval, and an agent refuses such an instruction if another agent passed it on.
Five operating levels, from normal running to "production is down", change what the agents may do. Deploys, anything that moves money, messages to real people, deletions and secrets need a person's approval at every level.
More than ten thousand automated tests check the code the agents write, and a contract test catches any change that would quietly break the version of the app already on people's phones. An agent may change a test, but it has to list every change openly, with what the test checked before and after, why it changed, and what would make the change the wrong call.
A development server can't send a real email, by design, and every production deploy is run by a person rather than an agent.
I also built the tools the agents work with: an operations dashboard, a voice channel they use to report and ask for decisions, and decision pages that are answered with a click.
I've written more about how I work with AI.
