I offer fixed-price services where the work can be scoped tightly, bespoke builds quoted against a written scope, and speaking and training engagements for teams who want to take their agentic coding to the next level.
Below are fifteen pieces of work that I have delivered as passion projects over the last six months.
Fifteen pieces of work since March 2026, each carrying its true state. These six lead, four of them with a case study behind the title. The rest of it holds the other nine, the migrations case study among them, labelled the same way.
Race management for sailing clubs, entry fee to published result. Live on the App Store in 22 languages at flyingstart.app.
A free iPhone app that tells you what has been added to or removed from your calendars since you last looked. Everything is compared on the device; no diary data leaves the phone. Download it on the App Store.
Idea to App StoreThe scoring engine behind Flying Start, checked against Sailwave, the software most clubs already use, so nobody has to take the results on trust: 144,539 comparisons drawn from 1,345 published results files, 99 clubs, ten countries. Method, scraper, engine and data are all published, so a club that doubts the match can rerun it.
Unmanaged risk to managed riskFour models from four vendors on the same diff. Three review it blind to each other, then defend or refute what the others found; the fourth reads the repository and rules on whatever is still contested. One model agreeing with itself is not evidence. The tool that runs it is open source.
AI-driven code and security reviewThe same journey written three ways and raced: a backend driver, Maestro, and the native engines, XCUITest and Espresso, across simulators, emulators and a handset. The times are not close, and knowing which engine to reach for beats picking one and hoping.
Automated test harnessesThe course version of how all of the above gets built, for founders, product owners and the people in an IT department who specify and sign off software rather than write it.
One person, one line of accountability. No bench, no juniors billed at my rate, no handover to somebody you have not met.
The problem is rarely that the leadership is unconvinced. It is that the engineers are, correctly, suspicious of a technology that mostly arrives as marketing, and no deck fixes that.
My first iPhone app took twenty-three minutes to build, and it did not work. I pasted a screenshot of the error back in and four minutes later it did. That number is genuinely astonishing and it is also the least useful thing I can tell your engineers, because twenty-seven minutes gets you a demo, and none of it gets you production. The talk is mostly about the gap between those two things.
Four things here are defined tightly enough to carry a price on a public page, so they do. Fixed fee, staged payments, and the estimate risk sitting with me, not you.
A working first version, and a written roadmap for everything after it: what to build next, what to leave alone, what the next number is likely to be. UK agencies quote £25,000 to £80,000 and spend the first fortnight writing a discovery document. There is also a five-day prototype at £6,000, roughly what an agency charges for the discovery document alone.
Four models over your codebase, three arguing and the fourth ruling on what they cannot settle, and a written verdict with the false positives thrown out. Code quality as well as security posture. It is not an accredited penetration test and does not pretend to be one; if that is what you need, you need a CREST firm and I will say so.
Two days with up to twelve engineers, hands on their own repository rather than a toy one, ending with the guard rails installed and running on real work. A single day is £2,750. Travel quoted separately, and outside the UK it is quoted before anything is agreed.
A conference keynote, an internal all-hands or a session for one team, with real repositories on screen. Most UK corporate bookings run £3,000 to £10,000, so this sits at the bottom of that and will not sit there forever. Remote is £2,000. Travel inside the UK is included; outside it is quoted before anything is agreed.
Fixed fee · staged payments · deliverables agreed in writing before the start · VAT not included
A fixed price on a vague brief is either padded to cover the unknowns or runs out of money halfway, and both end up being your problem. These are quoted against a written scope instead, still fixed once there is one.
Web, iOS, watchOS and Android, built and submitted. Priced once the scope is written down, which is what the roadmap above is for.
Priced off the content count and the integrations, both of which are countable before anyone commits to anything. Two done so far, one live.
The whole delivery, with your engineers alongside me so the capability does not leave when I do. Documentation is in the price, not a change request at the end.
The priced work is the front door to the quoted work, and deliberately so. A prototype, a roadmap or a review produces the document the larger quote is built from, so by the time there is a big number in front of you, both of us already know exactly what is inside it. None of it is a deposit and none of it is contingent on going further: if the roadmap says do not build this, that is what it will say, and you keep it either way.
From thirty months on what was billed at the time as the largest civilian IT programme in the world, down to four people in a shed. Retail, central and local government, healthcare, retail banking, life and pensions, supply chain, advertising and e-commerce, in the UK and across Europe, and regulated or politically loaded most of the way.
Tell me the problem rather than the technology.
Use the form. It reaches me directly and I reply within a working day.
Dom Brandreth. Working across the UK, Europe and remote, on site when the work needs it.
For an event, say the date, the audience and what the session needs to change. For a build, say what is in the way rather than what you want to use.