I like evidence more than opinions - including my own.

I'm Mark Sindoni Houghton. I'm a mobile product engineer: I build iOS and Flutter apps, and I own the decisions about what goes in them.

I've been writing native iOS since 2020 - first as a consultant at Infinity Works, then at Accenture after the acquisition, and since 2023 at Pocketworks, a UK mobile agency, where I'm now a Flutter Developer and Product Owner. That work spans health and fitness, transport, manufacturing and consumer products. The through-line is that I've rarely been handed a finished specification. Most of it has been ambiguous requirements, an existing app with problems nobody had named yet, or an early idea that needed shaping before it needed building.

Outside that, I build markrsh apps - a suite of products I research, design, build and ship on my own.

The part before the code

The thing I've become most disciplined about isn't Swift. It's the step before Swift.

Most of my ideas are wrong. That's not modesty, it's just the base rate, and the expensive version of being wrong is finding out after you've built it. So before I commit to anything now, I run it through research: four to six agents working in parallel, each interrogating a different angle - who else is in this market, what their users complain about, what people actually pay, what the category has collectively failed to notice.

Then the findings have to survive contact with evidence. Frequently they kill the idea, which is the point.

Where the habit came from

Before commercial software, I spent four years on a PhD in Computer Science at Leeds, building computational models of disordered fibre networks - work that produced two publications and a durable suspicion of anything asserted without a number behind it.

Research training gives you three things that turn out to matter more in product work than they have any right to: break an unclear problem into parts you can actually test, distrust your own first explanation, and be willing to publish the result that contradicts what you hoped for. The AI research workflow I use now is just a faster version of the same instinct.

What I'm useful for

I'm most useful to teams that need someone who can move between the strategy and the implementation - who'll argue about whether a feature should exist, then go and build it properly.

Concretely: native iOS in Swift and SwiftUI, with SwiftData, StoreKit and RevenueCat, App Intents and Siri, and Apple's on-device Foundation Models. Flutter and Dart for cross-platform work. Enough full-stack to build and run a multi-tenant web product on my own when a project needs one.

And the less technical half, which is usually what's actually missing: scope shaping, deciding what not to build, subscription and pricing strategy, analytics that answer a question rather than filling a dashboard, and getting something in front of real users early enough that their reaction can still change it.

The practical details

I'm British, and I live in Trento, in northern Italy. I work remotely with UK and European teams, on UK hours, and have done since 2023. English is my first language; my Italian is conversational and improving slowly.

I take on selected contract and consulting work alongside my role at Pocketworks. If you're weighing something up, here's how that usually works - or just tell me what you're building.