INAURO
Book online
(00) Software engineering · Stanmore, HA7 · UK-wide
Software rarely fails loudly. It drifts. We build the systems underneath, then stay long enough to watch them run.
(01) Who is answering the phone
A short account of why this company exists, written by the people who started it.
We met on a claims platform that moved about £40m a year and fell over most Tuesdays at two in the morning. Nobody could say why. It took eleven weeks to find the answer: a scheduled job written in 2016 by someone who had long since left, quietly retrying itself into the ground.
We fixed it in an afternoon. Then we watched the consultancy we worked for sell the same client a rebuild it did not need, and we understood that we had spent two years billing for confusion. That was the turn. We handed in our notice in the spring and set up on our own in Stanmore, four miles from the rack we had been nursing.
Inauro takes two projects at a time. The engineer who scopes your work writes the code, and you get their number and the repository on day one. There is no account layer, no delivery manager, no weekly status theatre.
We turn down anything we cannot staff that way, which is most of what we are asked to do. Much of this industry sells headcount by the month. We would rather hand over a finished thing that somebody else can read.
(02) The standard we work to
Anyone can ship. The bill for how it was built arrives in year two, and it is usually larger than the build.
Our one non-negotiable
Two weeks, fixed fee. We read the code you already have, sit with the people who use it daily, and hand back a written specification and a ranked risk list. Take it elsewhere if you want to — it is yours.
Internal tools, booking systems, client portals, operational dashboards. TypeScript and Postgres unless there is a reason not to, deployed to your infrastructure rather than ours.
One codebase, both stores. Store submission, review responses and the first point release are part of the price, because that is where most app projects quietly stall.
Getting two pieces of software to agree: payments, accounting, CRM, warehouse, the courier's endpoint that returns 200 on failure. Includes retries, reconciliation and an alert when it drifts.
The developer left, nothing is documented and the build has not run since March. We map what is actually there, get it compiling, and tell you plainly whether it is worth keeping.
Patching, monitoring, on-call inside working hours and a plain monthly note on what changed and what nearly broke. One month's notice, either way.
How a build actually runs
No stage depends on you signing up for the next one. That is deliberate, and it is why we scope first.
Two weeks, paid. Existing code read, users interviewed, specification and risk list written down.
Two-week increments, each one deployed to a staging URL you can open yourself. No demos from a laptop.
Repository, runbook and a recorded walkthrough. Ownership of the code transfers on final payment.
Optional. The alerts stay pointed at us for as long as you want them there, and not a month longer.
Increment 6 of 9 — staging deploy, Thursday evening
(04) Book online
Send the awkward version — the bit nobody wants to write down. We read every enquiry ourselves and reply within one working day.
