Minneapolis, Minnesota
Let us design the system that fits your firm.
Is there a spreadsheet three departments now depend on? A monthly subscription that still leaves a third of the work happening in email? A process that runs because one person knows the order of the steps?
Heuron builds the software that replaces it — at a fixed price, agreed before I start. One person, thirty years, and no surprises on the invoice.
Who I work with
Sound familiar
Three versions of the same problem
The spreadsheet that became a systemEvery firm, eventually
It started as a tracker. Now four people depend on it, one person actually understands it, and nobody will touch a formula without asking her first.
It works beautifully, right up until the week it doesn't.
The product that almost fitsPractice management · Case management · Project tools
You bought software for this. It does about seventy percent of what you need, and the remaining thirty happens in email, exports, and a folder on the shared drive that has a naming convention nobody follows.
You're paying a subscription for the privilege of still doing it by hand.
The work only one person can doMonth-end · Intake · Onboarding · The annual cycle
It runs because someone knows the sequence and what to check at each step. That knowledge isn't written down anywhere.
And that person is going to retire, or leave, or — far sooner — take two weeks off in July.
None of this is a technology problem. It's that you're the right size to have real process and the wrong size to have anyone whose job is software.
John Donaghy · Founder, Heuron Technology
Two ways in
A project, or a person on hand
Most engagements start here
Something built
You know roughly what's broken. We scope it properly, I price it once, and I build it. Weeks, not quarters — and you see working software along the way rather than a status report about working software.
You either run it yourself or I host it. Never locked in either way.
For firms without a technical anyone
Someone to think with
The vendor contract you can't evaluate. The security questionnaire a client just sent you. Build or buy. What to actually do about AI, as opposed to what everyone says about it.
A day or two a month, on retainer. Fractional CTO, if you want the term for it.
Same person, different arrangement. The second usually starts because of the first.
How a build works
Three stages. You can stop after the first and still come out ahead.
Discovery
Two weeks. I sit with the people who do the work today, then write it down: what the system has to do, what it connects to, what could go wrong, and what it will cost to build.
You own that document whether or not you hire me for the build. Take it to another firm and it will still be the best specification anyone there has ever been handed.
The build, at a fixed price
Priced from the specification, agreed in writing, before a line of code is written. If it takes twice as long as I expected, the price doesn't change. That risk is mine, and carrying it is a real part of what you're paying for.
Then you run it, or I do
Either way, you already own it. The source lives in your own source repositories and the infrastructure in your cloud subscription from the first week — not handed over at the end, but yours while the work is happening.
So hosting is a convenience you buy, not a dependency you acquire. There's no version of this where you're stuck with me, and that's in the contract rather than promised on a website.
No mystery
What it costs
Build prices are fixed and agreed in writing before work starts. Ranges narrow to a single number at the end of discovery.
People ask why the prices are on the website. Because you were going to ask in the first meeting, and neither of us needs to spend a meeting on it.
Why me
Half my career in startups. The other half inside very large companies.
The startup half
Twenty years of building products where shipping was the whole job — market data, healthcare, enterprise software, the New York Stock Exchange. Then five years running Axon Financial Systems, a company I co-founded, self-funded, and sold in 2019.
That's the half that lets me move quickly and quote a fixed price without flinching.
The enterprise half
Since 2019, software and platform engineering inside large organizations — first RSM US, a national accounting and consulting firm, then a national title insurer, building cloud infrastructure under genuine security, audit, and change-control constraints.
That's the half that means nothing I hand you is a prototype, and that I take your clients' data as seriously as you have to.
Firms your size need both halves and can't usually buy either one.
The questions you're too polite to ask
How does one person do this — and what happens if I get hit by a bus?
The first is about capability. A system like this once meant four developers and most of a year; it now takes weeks, because I've spent the past six months building the tooling that makes it possible — infrastructure that runs AI coding agents at volume behind enforced test, review, and merge gates, so work moves fast without anyone losing track of what changed or why.1 That's what lets a one-person practice quote a fixed price and mean it.
What the tooling can't do is decide what to build, notice the requirement nobody thought to mention, or take responsibility when something turns out to be wrong. That part is still a person, and here it's the same person throughout — in discovery, during the build, and on the phone eighteen months later.
The second question is the fair one, and most people are too polite to ask it out loud. The answer is structural rather than reassuring.
You own everything from the first week. Not at handover — from the start.
The source sits in your own source repositories. The infrastructure sits in your own cloud subscription. The tests, the deployment automation, the documentation, the keys — all in your possession while the work is still going on. So the worst case isn't that your system dies with me. It's that you hire someone to pick up a documented, tested, running codebase — and that is an ordinary thing to hire for.
Compare that honestly to the alternative. An agency's continuity is a promise about staffing — and when their senior engineer leaves, which happens, you find that nobody remaining understands your system either, and you hold no source, no deployment, and no leverage. Mine is a promise about what is already sitting in your accounts.