About 10+ years Open to work
I am a full-stack developer. Most of the last decade has been spent in Python — Django underneath, Wagtail on top of it more often than not — and the rest in whatever the front of the application needed to be that year.
The work I get asked for has a shape to it. Someone has a system that matters to their business and something about it has stopped being safe to touch: the version is old, the person who built it has gone, or it was never quite finished. I am usually the second developer on a project rather than the first, and I have come to think that is the more interesting job.
I
Ten years is not ten years of skill.
It is ten years of consequences.
You learn what you have to live with.
II
what a decade of this work teaches
Observation 01
The first week should produce almost nothing.
Every expensive mistake I have made was made confidently and early, before I understood the system I was changing. The weeks that look unproductive from outside — reading code, reading data, asking what the business actually does — are the ones that decide whether the rest goes well.
Observation 02
Everything else is negotiable.
Interfaces get redesigned, frameworks get replaced, teams change. The shape of the data outlives all of it, and a model that does not match what the business does will quietly tax every feature built on top of it for years.
Observation 03
Interesting infrastructure is paid for weekly.
Clever architectures can cost more in on-call hours than they ever save in compute. Unless there is a specific problem that demands it, the answer is the well-understood thing that the next developer will recognise on sight.
Observation 04
Documentation written afterwards is fiction.
Not a wiki nobody updates — the reasoning, in the repository, next to the thing it explains. Six months later I am as much a stranger to the code as anyone else, and the note I left is what stops the same question being answered twice.
Observation 05
Being needed is not the same as being good.
A developer nobody can replace is a risk their client is carrying, whether or not anyone has said so. I document, hand over, and would rather be kept because the work is useful than because leaving would be expensive.
III
the projects I’m the wrong hire for
An odd thing to put on a page meant to win work — but the alternative is finding out three weeks in, which is worse for both of us.
I build, and I care about type and spacing. I am not a substitute for a designer on identity work, and I will say so rather than take the fee.
iOS and Android as first-class native applications are not my field. Web, and web that behaves properly on a phone, are.
I can integrate a model somebody else trained. I am not the person to choose or train one.
I work best alone or alongside a small team. On a project with twenty developers I would be the wrong shape of hire.
Send a note
If none of that put you off, tell me what you are working on. I will tell you honestly whether I am the right fit.