The Idea
Akshay Narisetti had been building robots since he was 12. By 23 he had built more than a hundred of them and become a founding member of Omi, one of the world's largest open-source AI wearables. Google noticed. Its VR team came calling, and Narisetti turned them down. Three times.
What he wanted to build instead kept nagging at him and his co-founder Gabriel Dymowski, the former CEO of blockchain startup DoxyChain. Every important conversation in their lives happened in person: investor meetings, doctor visits, client calls, team standups. And every notetaking tool on the planet only worked over Zoom. For a real-world conversation, the best technology available was still a pen.
This was the same category that had just buried Humane and Rabbit. Hundreds of millions of dollars, glossy launches, dead on arrival. Narisetti's read was that the hardware was never the problem. The use case was. AI needs context to be useful, and most of that context is spoken out loud, offline, where no app could reach it.
So they built a credit card-shaped puck that sticks to the back of your phone, records both sides of a conversation, and hands you the transcript. This was the beginning of Pocket.
The Execution
The lesson?
Pocket's idea was not novel, and Narisetti will tell you so himself. Humane and Rabbit raised hundreds of millions on flashier pitches and still cratered. What Pocket had was a founder who had already shipped open-source hardware and knew the Shenzhen supply chains that quietly kill most gadget startups before unit one. The concept was commodity. The execution was not. In AI hardware, the graveyard is full of better demos than Pocket's. It just shipped first.