Claude Code · React · TypeScript · Electron · Remotion · GitHub · Chrome Web Store

Claude Code builds, held to written checks.

My own meeting-notes app, fact-checked AI tool reviews for a client, 14 websites on automatic deploys and a browser extension on the Chrome Web Store. Each one is below, with the checks it had to pass.

How I work with Claude Code.

Claude Code is my engineering partner. I set the rules and write the checks, and the work is not done until it passes them.

On the fact-checked AI tool reviews I build for a client, that rule is enforced in code. Since 17 October 2026, a Claude Code hook checks each commit Claude makes against the parts of the written standard a script can test. If a finished page fails, or one of its quoted facts cannot be re-found, the commit is blocked.

Read the case study

What I have built with Claude Code.

ThisIsNoted, my own meeting-notes app

AI notes you can check. Every AI point shows where it came from, and plain code (not the AI) flags names, numbers and dates that are not in your notes, the meeting invite or what was said on the call. Your typed text is never changed.

1,327 automated tests and a 187-step click-through of the real app were passing on 26 October 2026. There is also an evaluation kit of 11 made-up meetings with answer keys.

Built with: Claude Code, Electron, React and TypeScript.
Status: My own product, not released yet. The Windows version comes first.

A ThisIsNoted note with its draft: main points, a decision and an open question, each quoting the typed line it came from, then action items with who and when. — Each point quotes the line it came from. Shown in practice mode (simple word rules, not real AI) with sample notes.

See it on the homepage

Fact-checked AI tool reviews

A Claude Code project that tests AI tools hands-on, and will not let a review page out until every vendor fact in it is re-found in a saved copy of the vendor’s page.

Every vendor page a review relies on is saved. A script re-finds each quoted fact in the saved page and records the page, the time and the result. So far that is 4 review pages and 208 facts, all re-found, over 175 commits between 16 October and 1 November 2026.

For: A client building a US review site about AI tools.
Status: Client build, in delivery.

Read the case study

14 websites on automatic deploys

I moved 13 sites from drag-and-drop uploads to GitHub with automatic deploys, and launched a 14th the same way. Logged test deploys went live in 7 to 14 seconds.

Every move followed a written checklist (live status, security headers, deploy source, hardcoded paths, assets, domain), with the results logged. All 14 sites are set up with 5 security headers.

The hand-over is a one-page guide, written so no developer background is needed. In 4 steps, it teaches the team to change their own sites by asking Claude Code.

For: A US membership business that teaches business owners to use AI.
Status: Delivered, May 2026.

See it on the homepage

Cart-filling browser extension

It loads a used-book seller’s morning picks into his shop baskets. After he pays, it reads his orders and logs each one into his inventory system. It never buys, never sees a card and stores no password: it works inside his own signed-in browser.

Tested live: baskets filled on both shops, and 3 real orders from one shop logged, with 13 duplicates refused. It does not guess: an order it cannot match without doubt is refused, with the reason.

Status: Built and published on the Chrome Web Store.

See it on the homepage

A launch video, made from code

A 36-second launch video for a made-up app, built entirely in code with Remotion and Claude Code. There is no editing app and no stock footage, and the music is generated by code on the same beat as the picture.

Built with: Remotion, React and Claude Code, in wide and vertical versions.

A still from the launch video: a made-up meeting-notes app recording a call, with two action items pulled from what was said and the line “Every word, captured.” — A still from the 36-second launch video, for a made-up app.

Watch it on the homepage

Checks on other builds

All 45 commits on a seller dashboard were written with Claude Code.
Read the case study
On a used-book sourcing engine, 600 checks (as of November 2026) run before any change ships.
Read the case study
An Instagram AI chat agent I built for an agency had 219 automated checks across 7 test suites. That engagement ended in July 2026.
Read the case study

After handoff.

A finished build runs on your accounts, with full documentation and a recorded walkthrough. It is yours to keep.

If your team wants to extend or fix it without me, I can run paid training sessions.

If you would like me to keep looking after it, the Maintenance and Growth plan is $200 a month and covers up to 5 hours of updates, monitoring and small fixes. Bigger ongoing work is priced per scope.

Price and how a build goes.

You get a written scope and a fixed price for each milestone. You sign off before I build anything, so there are no surprise invoices.

Your hosting, domain costs and third-party API spend are not part of the price. They are listed in the scope, and I don’t mark them up. For a smaller job, hourly consults and fixes start from $15 an hour.

See the price list

How a build goes.

Day 1
Call and audit
A 30-minute call. I look at your current setup, ask where the hours go, and point out what is worth automating first.
Days 2 to 3
Written scope
Deliverables, timeline and a fixed price per milestone, in writing. You sign off before I start.
1 to 3 weeks
Build
I work in your tools, so the handoff is clean. You see working pieces every few days, with a weekly check-in.
Final week
Handoff
It runs on your accounts, with full documentation and a recorded walkthrough. Yours to keep.

Want something built with Claude Code?

Tell me what it has to do and how you would check that it works. We’ll start with a 30-minute call about your current setup and where the hours go.

Book a call See the Claude Code builds