About
Built at the intersection, on purpose.
Technology strategy, engineering, AI and governance used to sit in separate conversations. AI has moved faster than most organizations' ability to govern it, and that has pulled the four together whether anyone planned for it or not. We work across all of them as one line of work.
Technology decisions rarely sit in isolation.
A new product affects operations. An AI system affects data, people and risk. A platform decision affects how an organization works for years.
The industry is mostly organized around one of those at a time: strategy, or engineering, or compliance. There's good reason for it. Each one is deep enough to be a career, and specialists genuinely earn their keep. It's just that a technology decision now usually carries an AI dimension, and an AI decision usually carries a governance one. When those sit with separate teams working from separate vocabularies, the hardest part of the project quietly becomes the handover between them.
So we try to understand the whole picture before recommending a direction, and we keep the decisions that really matter out in the open where you can weigh them, instead of letting them get settled quietly inside a delivery plan.
Think. Build. Govern.
Three disciplines, one connected approach, usually with the same people from the first conversation through to the thing running in production.
Think
Understand the business context, then make the decisions that are expensive to reverse, before they get made by default.
Build
Turn that direction into products, platforms, integrations and AI systems that hold up after launch.
Govern
Put the ownership, controls and visibility in place to run it responsibly as it scales.
How we think about the work.
There's a lot of pressure to adopt AI at the moment, and much of it is well founded. But not every problem that arrives labelled “AI” turns out to be one. Often what's really there is a data quality issue, a process nobody has fully agreed on, or a decision that's been waiting a while for someone to make it. We'd rather spend the first conversation working out which of those you're looking at. If AI genuinely is the answer, you get there with a much clearer brief. If it isn't, you've saved a build.
Modernization works in a similar way. Teams living with an ageing platform feel the friction every day, and that frustration is real and worth taking seriously. It's still worth separating two situations: a system that is simply old, and a system that has become the limit on what the business can do next. The second usually justifies significant investment. The first often needs something smaller and less disruptive, and we'll say so, even when the bigger rebuild would have been the larger piece of work for us.
Governance is where we see the most goodwill go to waste. Frameworks like the EU AI Act, NIST's AI RMF and ISO/IEC 42001 describe outcomes rather than implementations, so it's genuinely hard to know what to build, and what often results is a careful policy document that the delivery team never quite touches. That isn't a failure of intent, it's a translation problem. Translating it is the part we enjoy: risk classification that reflects how your systems actually work, controls that live in the product, and monitoring that somebody owns.
We run what we recommend.
We build and operate our own products in production, isitholiday.com and mypersonalqr.com, with real users and on infrastructure we maintain ourselves. That's deliberate. It's the difference between recommending an architecture and living with one: you find out what actually breaks, what costs more than the estimate said, and which decisions you regret six months in.
Want to work together?
Tell us what you're trying to achieve. We'll help you understand the options and identify a practical next step.
Start a conversation