Before Mellmo #

I was already deep in the Apple ecosystem before this chapter started. I’d been building for Apple platforms since Mac OS X Dashboard widgets, shipped one of the earliest iPhone apps through Installer.app on iOS 1 (before the App Store existed), and had PegJump in the App Store on day one — July 2008. Dashbuster was there too. PegJump went on to rack up millions of downloads and is still live today. Its icon hung in the window of the Apple Store on Fifth Avenue. That’s the context I brought to Mellmo: not someone learning iOS, but someone who’d been building on Apple’s platforms since before most people took them seriously.

The work #

Roambi was one of the first apps to prove that the iPad wasn’t a toy — it was a new surface for serious business data. We built beautiful, interactive data visualizations for mobile at a time when most enterprise software looked like a spreadsheet crammed into a smaller screen.

This was a genuine platform transition moment. The iPad launched and within months, executives wanted their dashboards on it. Roambi was the answer — and Apple noticed. Our work was featured prominently in the App Store, on WWDC banners, and in Apple’s own marketing of what the platform could do. Roambi received iPad App of the Year honors — not because it was pretty (though it was), but because it showed what the hardware could actually do when you built for it natively instead of porting a desktop UI.

What I did #

I led iOS development for the flagship visualizations — SuperList, Elements, and Layers. This was deep, performance-sensitive Objective-C work: custom Core Graphics rendering pipelines, gesture recognition systems built before UIKit had good abstractions for them, and a data pipeline architecture designed for offline-first mobile BI. We were inventing interaction patterns for tablet-based data exploration because the platform was too new to have established ones.

When Apple announced the iPad, we got pre-release hardware and had a matter of weeks to deliver a product that would be ready for the launch wave. That kind of constraint sharpens your instincts fast — you learn what matters, what’s good enough, and what needs to be perfect.

I also co-invented the patent for Displaying Table Data in a Limited Display Area (US20110289397) — the method for rendering large data tables on mobile with dynamic, interactive visualization. It was the technical core of how Roambi made complex data feel natural on a 10-inch screen.

Forbes profiled the Mellmo culture in 2010, with me as the centerpiece example — a senior software developer who was also a professional jet skier. The piece was about hiring people who take risks and push boundaries. It wasn’t a bad description of what the engineering team was like.

Why it matters #

Roambi taught me what it looks like to ride a platform wave well. The people who win during platform shifts aren’t the ones with the best resumes — they’re the ones who see the new affordance clearly and move fast. That lesson has shaped everything I’ve done since.

It also set the pattern for the rest of my career: take a new surface, understand what it makes possible that wasn’t possible before, and build something that earns the platform’s attention. Apple featured us because we respected what the hardware could do. That instinct — build for the platform, not on it — has informed every technical decision I’ve made since.

More context: The Mellmo Days, A Decade of WWDC, and Why The Best Developers Do Backflips on Dabble or Die.