Digital quality inspection
Inspection software that replaces the paper checklist, with pass and fail logic against measured values.
Services
Twenty-five pages, in six groups. Every one says plainly what it covers, where we have shipped, and where we have not. If a product already does what you need, that is on the page too.
Book a 30-minute scoping call See all twenty-five
An engineer reads what you send, not a sales rep. Answer within one working day.
| Digital quality inspection Readings, checks, certificates | 3 |
|---|---|
| Building software Web, mobile, product, devices, design | 6 |
| Data and AI Pipelines, models, documents, fleets | 4 |
| Infrastructure and security Cloud, delivery, controls, identity | 4 |
| Testing and ways of working QA plus six engagement pages | 8 |
Everything we do
Nothing is hidden behind a contact form. Each page names the work, the technology and the cases behind it, and says when we are the wrong choice.
Where a reading, a check or a certificate has to hold up months later.
Inspection software that replaces the paper checklist, with pass and fail logic against measured values.
Readings taken from torque wrenches, pull testers and gauges over Bluetooth or a vendor SDK, calibration checked at capture.
Statutory lift examinations under LOLER or ASME A17.1, offline capture in a shaft, certificates in your own format.
The systems a business runs on, and the interfaces people use to run them.
What to build when no product covers your process, and when to buy one instead.
Angular and React over .NET or Node, deployed into your own cloud tenant, including legacy modernisation.
Field apps for gloves, bad light and no signal, reading instruments over Bluetooth.
Building software you sell: tenancy, tiers, self-service sign-up, white labelling and self-hosted delivery.
The platform devices report into, over-the-air firmware delivery, and readings pulled from instruments.
Interfaces that carry work, with prototypes tested before production code and a component library handed over.
The unglamorous layer underneath, and an honest read on whether a model is worth building at all.
Ingestion, storage, movement and migration, plus the monitoring that tells you a feed has stopped.
A two-week feasibility read first. Half of them end with advice not to build a model.
Retrieval over your own documents, with citations and permissions, proved on a two-week pilot.
Three reasons to connect equipment and three different builds. Pick before you buy sensors.
Who owns the account, who is allowed to deploy, and who is allowed to see what.
Azure and AWS in accounts held in your name, with the pipelines and the exit written down at the start.
Pipelines, environments, monitoring and a rollback somebody has actually rehearsed.
We build to the controls a framework asks for. We do not issue certificates, and we name the firms that do.
Access enforced at the server, tenant isolation, and entitlements that cannot be switched on from a device.
Automation and regression handed over with documentation, rather than a suite only we can maintain.
Offshore QA outsourcing, automation, regression and release testing, with the test plan written alongside your people.
Five engagement models plus support. They differ in one thing: how much management you hand over.
The hub comparing all five models, with the situation each one suits and where it stops working.
A long-term engineering site under your standards, where we carry recruitment, HR, payroll and equipment.
A team that works only on your product, with a lead on our side running delivery.
Engineers inside your process, under your management, when you have capacity to direct them.
One price for a scope that is genuinely settled, with the overrun carried by us.
A Vietnam team you will own, with the transfer price agreed in the first contract.
For the system whose original developer has gone, and which still has years left in it.
Working from the United Kingdom? The UK page covers what BS 8539, LOLER and a UKAS calibration certificate actually demand of the record, where your data sits, and the two British builds behind it.
Still choosing
Nobody arrives knowing which of the twenty-five they need. Most people know one of these three things instead, and any of them is enough to start a useful conversation.
Three fields. An engineer reads it, not a sales rep. You get an answer within one working day.
Our work
We reference the same engagements across the site because they are the ones we can show. A longer list would be a list of logos.
Who does the work
Small by outsourcing standards, and deliberately so. Twenty-five service pages does not mean twenty-five teams. It means one team that has shipped across these areas and will tell you which ones it has not.
You interview everyone we put forward, on every engagement model. We screen and shortlist, you take the final call.
How we work
The models differ in one thing: how much of the management you hand over. Everything else, including who owns the code, is the same in all five.
Changing model later is normal. Augmentation into a dedicated team is the common direction, and the people stay.
Engineers join your team and work inside your process, your board and your code review.
A team that works only on your product, with a lead on our side running delivery.
A long-term engineering site under your standards, where we carry recruitment, HR, payroll and equipment.
04
One price for a scope that is genuinely settled, with the overrun carried by us.
Everything in the centre model, with the whole team moving into your own Vietnamese entity on an agreed date.
FAQs
It is a fair question and the honest answer is that these are not twenty-five separate practices. They are one team writing down what it has shipped, split by what a buyer searches for. Several pages say plainly that we have no published case in that area, the AI pages among them. Read those first if you want to test whether the rest is honest.
Start with the pillar page of the group that fits best: digital quality inspection, custom software development, software QA and testing, or the offshore development centre. Each one covers the group and links down to the narrower pages.
No, but we only claim the ones we have shipped in. Safety testing equipment, education, automotive compliance, construction and work at height, and industrial IoT are where our cases come from. If your sector is not in that list we will say so rather than pretend the experience transfers automatically.
Ho Chi Minh City. A UK team gets a morning overlap, an Australian team gets most of the day. A US west coast overlap has to be engineered rather than assumed, and we agree it in writing before anyone starts rather than discovering it in month two.
Mostly small and mid-sized companies where one person owns the decision, and product teams inside larger firms. The smallest sensible start is one engineer under augmentation or a lead plus two or three under a dedicated team. Both of our largest engagements began around that size.
Yes, and we prefer it. A feasibility read, a two-week pilot, one feed end to end, one building running. Every service page has a first step that is small enough to stop after, and that is deliberate.
Yes, and two of our largest engagements began that way. We read the code first and tell you what is salvageable before quoting, because quoting a rescue without reading it is how the first supplier got into trouble.
Your company, from the first commit, in every engagement model on this site. Repositories, pipelines and documentation are handed over on request rather than at the end of the relationship.
You do not have to know which of the twenty-five pages you need. Tell us what happens today, who it frustrates and what it costs, and thirty minutes is usually enough to name the shape of the answer.