Every era of computing was defined by its interface.
The command line gave way to the mouse and the desktop, and computers went from tools for experts to tools for everyone. The touchscreen put the internet in a pocket and changed how billions of people live. Each time, the machine underneath mattered less than the way we reached it. The interface decided what was possible, who it was for, and what it felt like to use.
I started interfaces.inc because we are living through the biggest shift yet. Computers can now think. But we still talk to this new intelligence through interfaces built for the old world, a text box, an app, a screen you go to and type into. The intelligence has leapt forward. The interface has not. That gap is the most important open problem in computing right now, and it is the one we work on.
I have been chasing this for the last four years.
When modern AI arrived, I could not stop building at the edge of it, and the question underneath everything I made was the same: how is a human actually supposed to talk to this? I built memory into language models before ChatGPT existed, because an intelligence that forgets you felt broken. I built one of the first open-source AI wearables, so it could hear the real world, and it grew into Omi, now used by hundreds of thousands of people. I built agents that use a computer the way a person does, so it could act, not just answer. Each was a different attempt at the same problem: the interface between people and intelligence.
Silent Speech is the clearest shot I have had at it. Voice is becoming the way we interact with AI, faster and more natural than typing. But you cannot talk out loud most of your day, in an office, a meeting, a train. So I built the way to do it silently, using only how your mouth moves, on the devices you already own. It is live today, and it is the first piece of something much larger.
A bit of background.
I have been building things people use for as long as I can remember. At thirteen, in Israel, I made mobile games, twenty of them, over a million and a half downloads, number one in the app store, all on my own. I learned early what it feels like to put something in front of real people and watch how they use it.
Then Israeli military intelligence, where I went as deep on the technical stack as you can go, hardware and firmware up through software, in one of the top units. (That is also where I first saw, firsthand, how much the interface shapes everything: they took away our computer mice to force us onto the keyboard, and the same machines, in the same hands, became something else entirely.)
After the army I was a founding engineer at a cybersecurity company, where I helped take a raw idea to a real product used by real customers, through to its acquisition by Darktrace. I learned how to ship things people depend on, not just build demos.
That is the through-line I would actually claim: I build, I go deep, and I ship. interfaces.inc is where all of it points now.
What we believe.
Intelligence is no longer the hard part. The question that defines the next decade is how we live with it, and we believe the answer is presence: AI that dissolves into your world instead of sitting behind a screen, that is with you, aware of your life, reachable as naturally as a thought. The interface, not the model, is what turns intelligence into something that changes how you live.
That is what interfaces.inc exists to build. Silent Speech is our first step.