I told people in college that I wanted to build an operating system unlike anything that existed. Not a faster Windows, not a cleaner Mac — something fundamentally different. They laughed. I didn't have the language for what I meant yet, but the image was clear: an OS that touched every part of your life, adapted to how you actually worked, and got out of your way. AI is the first technology I've encountered that actually makes that real.
🗣️ The Moment I First Understood What Interfaces Could Be
It started before I even had the vocabulary for it. I was learning to use text-to-speech on my computer because someone close to me had special needs, and something clicked in a way I wasn't expecting. I thought — wait, what if I could just tell my computer to go to the next page? What if the device just listened? I didn't know if that was technically possible at the time. It didn't matter. I couldn't stop thinking about it, and that one moment planted something I've been building toward in every app I design since.
My whole design philosophy now starts from one question: where is the friction, and how do I remove it? If someone has to manually type something into an app I built when they could have just said it out loud, I feel like I've already missed the point. Every app I'm building right now — Sentio, Imori, the whole Namos Labs stack — is oriented around voice-first capture, because that's how people actually think. Out loud. On the move. In the middle of doing something else entirely. The keyboard was never the natural interface. It was just the one we had.
🌐 Screens Are Getting Quieter
The shift I've been watching isn't that interfaces are getting more powerful in a flashy, feature-heavy way. They're getting quieter. Fewer buttons, less noise, more of a relationship between you and the device where context is understood rather than entered. You're reading something, you glance away, and it knows you're done. You say "show me a simpler version of this" and it does. You speak what you need — a bill paid, a task logged, a thought you don't want to lose — and the system handles the rest without you touching a single field.
That's not science fiction. The individual pieces already exist. What's missing is someone designing the whole system with that intention from the very beginning, rather than bolting voice and context onto an interface that was originally built for mouse clicks and form fields. The OS I imagined in college wasn't about raw capability. It was about intentionality — building something that works the way humans actually work, rather than asking humans to adapt to what was technically convenient in 1995. Real work doesn't happen in perfect, uninterrupted sessions. It involves pauses, retries, forgotten context, and changing intent. The next interface has to be built with that reality in mind, not around a fictional version of a consistent, always-available user.
What I keep coming back to in my own builds is that the goal isn't to impress anyone. It's to disappear. The best interface is the one you stop noticing because it's just doing what you meant. That's the bar I hold myself to, and it's a much harder bar to clear than adding features.
♿ Accessibility Has to Be the Default, Not the Afterthought
The version of this future I care about most is the one that's accessible by default. A voice-first, context-aware interface isn't just more convenient for most people — it's more inclusive than anything we have now, because it's built around how humans naturally communicate rather than around what happened to work when the first graphical interfaces were designed. Someone who is blind shouldn't need a workaround to use a well-designed app. Someone who is deaf shouldn't get a degraded experience. Someone whose hands shake, whose eyes are tired, whose brain works differently — all of them should be served by the default, not by an accessibility menu buried three layers deep.
This connects directly to what I've been writing about in my 2026 predictions: the tools that matter now are not the most impressive demos, but the ones that persist across time, handle failure gracefully, and reduce cognitive load instead of adding to it. Smaller, focused systems that actually fit how people work are going to win. Not the biggest models. Not the flashiest launches. The most dependable ones — software shaped like human work rather than machine work.
🤖 What AI Actually Makes Possible Here
What's genuinely different now is that AI can hold context across a whole interaction. It's not just responding to a command — it's tracking what you're trying to accomplish, filling in the gaps you didn't know to fill, and adapting in real time as your needs shift. That's what makes the operating system I always imagined actually buildable. Not a tool you use, but a system that helps you run your life — your work, your finances, your health, your creative output — with the lowest possible friction between a thought and an action.
The teams that win in this next phase won't be the ones shipping the most features. They'll be the ones building systems that disappear into the workflow, persist across time, and treat reliability as the core product rather than a footnote. AI stops being a destination you visit and becomes infrastructure you run on. That's the shift. And once you see it that way, the question stops being "what can AI do?" and starts being "what should it quietly handle so I can focus on everything else?"
I've been thinking about this since college. I'm just glad the tools finally caught up to the idea.
What's the one part of your current workflow that you think should just stop requiring human input altogether?
