What do you need from us?

Alongside, nearly always. Most engagements have meant slotting into a team that already exists and already has its own way of working — covering the ITV Hub+ app through a permanent-hire gap, working next to an in-house iOS team at Fanatix, mentoring developers at Tesco and Tignum, and at Hotels.com passing technical knowledge across to the Mobile Web, Android, BlackBerry, Nokia and Windows teams.

We're used to working with your designers, your server team, your product owner and your QA rather than around them. If what you need instead is a self-contained build handed back finished, that works too — say which you want.

Yes, and it's the most common way we start. Decades of walking into other people's code mid-flight means we're just as often called in to fix what everyone else has given up on as to build from a blank page — restructuring the TVPlayer iOS and tvOS apps, rewriting a Lloyds banking codebase for two project teams, resolving critical memory and JSON faults in a property viewer, and inheriting the Vox Cinema apps and then maintaining and developing them for two years.

What you'll hear from a lot of developers and agencies is that the codebase is a mess, that the answer is to rewrite it, and that the new one will be better. We wish that were true. Usually a rewrite just moves the same problems further down the line — the decisions that made the old code awkward get made again, under deadline pressure, by people who weren't there the first time. Two years later someone is saying the same thing about the replacement.

Sometimes a rewrite genuinely is the right call, and we'll tell you when it is. But it's the most expensive option on the table, so it should have to earn that — which means understanding what is actually wrong before anyone reaches for a blank page.

You don't need to tidy it up first. Seeing it as it actually is tells us more than seeing it cleaned up for us.

From a few days to as long as you want. TVApp — a cross-platform media client for tvOS, iPad, macOS and Android — was built and shipped in ten days. A storybook app for a children's charity was twelve weeks. TVPlayer ran ten months, BlackBerry twelve, and leading the mobile division at dnata Travel and Travel Republic ran eighteen months across three extensions. The Vox Cinema apps ran longest of all: two years of continuous support and development on a codebase we inherited.

The shape follows the problem rather than a standard package. A diagnosis on a stuck project can be a few days; standing in as technical leadership is ongoing by nature.

Yes, and that's the point rather than a limitation. We don't sell headcount. The premise of the practice is that two deeply experienced people with modern AI tooling now deliver what used to take a squad — so the people you talk to are the people doing the work, and nothing is passed down to someone more junior once the contract is signed.

Where a project genuinely needs more hands, we'll help you source them, and hiring is something we do rather than something we hand off — we've sifted funnels of a thousand applicants and know what a good mid-level engineer looks like on paper and in a room.

Leaving a competent team behind is often how an engagement ends. On Travel Republic the client's own account is that the project landed on time and then the foundations were built for the team now running it; at Hotels.com, twelve months in, the ask changed from building the iPad app to organising and managing the entire mobile division. The aim is that you don't need us indefinitely.

It does mean we take on a small number of engagements at a time. If we can't do yours properly, we'd rather say so than stretch.

Both, in-house, natively. Dominic's depth is on Apple platforms — iOS, iPadOS, tvOS, watchOS and macOS since iOS 3 — and Alex is expert on native Android and iOS, including running split codebases across both. TVApp is SwiftUI on Apple and Kotlin on Android, not a cross-platform shortcut.

Then it's probably still in range. We work on both sides of the wire — microservices, RESTful APIs, event streaming, caching and the data layer — as well as AI systems and LLM integration, robotics, IoT and embedded work, and taking a finished product to market. The capability map above is the honest version of what we're deep in; if your problem sits outside it, we'll tell you rather than learn on your budget.

A conversation, before anything else — a chance to understand the problem properly, and to find out how you want us to work. Some teams want us in their standups and their repo from day one; others would rather be left alone until there is something to look at. Both are fine, but we would rather ask than assume.

What we don't insist on is an upfront specification. If you don't have one, it doesn't matter — we're used to working with ideas and turning them into reality.