Three ways to work with us.
Software you cannot buy off the shelf. A development team that wants AI without the risk. An old system that has to change. Each one opens with the problem, then what we do about it, then what you get.
Product development
Software that fits your business, built in weeks and kept moving.
The problem: nothing off the shelf does what your business does. The custom quotes you have had run to hundreds of thousands, months before you see a screen, and a maintenance bill every year after. So the spreadsheets stay, and the double entry with them.
What we do: you bring the business and the problem. We bring a small team, an AI engineering team and a human lead, working from a written spec you approve. We build in isolated branches with tests and documentation, and release through two review gates and your sign-off. You see it on staging before your staff do.
What you get: working software in weeks, then a steady stream of releases as the business changes. After launch we keep running it: releases, support knowledge, monitoring, the lot. Or we hand it over to your team with the briefs and the rails intact.
Includes
- Discovery and written specification
- Build with a specialist agent team under a human lead
- Tests, help content and release notes with every change
- Three environments, two review gates, blue/green production
- Ongoing operation or a documented handover
Good fit
Businesses with a real operational problem and no off-the-shelf answer. Owners who want software they can run on for years, not a prototype.
AI engineering enablement
Your developers, working with AI, without it getting loose in production.
The problem: your developers have tried AI coding tools and the results are mixed. Fast on day one, messy by week three. Nothing is governed, nothing is reviewed the same way twice, and you cannot put your name to shipping it. The missing piece is the team structure around the model.
What we do: we bring the structure. Agent roles with an ownership table for your codebase. Onboarding briefs for each role, written with your engineers. Guardrail hooks tuned to your rules. A pipeline with review gates and a human approval. Handover and memory discipline so the team improves instead of repeating itself. Then we work alongside your developers on real tickets until the method is theirs.
What you get: a development team that ships faster and is safer to trust, a written process your people own, and no ongoing dependency on us.
Includes
- Codebase and workflow assessment
- Role design and ownership table
- Onboarding briefs per role
- Guardrail hooks and scoped connectors
- Pipeline, review gates and release process
- Working sessions with your engineers on live work
Good fit
Software teams of any size who want the speed of AI development with the safety of a proper engineering process. Especially teams that have been burned by the fast-then-messy pattern.
Modernisation and integration
Off the old system, with your history intact and no big bang.
The problem: the business runs on desktop software written a long time ago, often by one person who has since moved on. It works, until it does not. Every report is a favour. It cannot talk to Xero, the website or a phone, so the same order is keyed in three times. Nobody dares touch it, and a big-bang replacement is a risk you cannot take.
What we do: we move it to the web without losing what made it work. We read the old data model properly, migrate the history rather than starting from zero, replace it module by module where that is safer, and wire the result into the services the business already uses: Xero for accounting, Stripe for payments, Make.com for the rest.
What you get: a modern web system your staff can use on the floor and on the phone, your years of jobs and customers still in it, and a cutover you can rehearse until it is boring. We have done this for the print, signage and display industry with BLAQ. The same method applies to any operational system.
Includes
- Legacy system and data model review
- Web replacement, module by module where needed
- Data migration with validation, re-runnable until cut-over
- Xero, Stripe and Make.com integrations
- Parallel running and cut-over planning
Good fit
Businesses whose core system is old, single-vendor or single-developer, and who need it modern without a risky big bang.
One method under all three.
Whichever you choose, the work runs through the same team model, the same guardrails and the same pipeline we use on our own product. You can read all of it on How we work, and see it applied on BLAQ.
Contact
Tell us what you are running on.
Spreadsheets, a desktop system from another decade, or a team that wants to build with AI without betting the business on it. Start with a short conversation.