I'm in San Francisco. The decision took about five minutes once I actually made it, even though I'd been circling it for much longer.
The honest version: I spent the last two years building a social app in Turkey and the US, learning what it means to be responsible for a technical product end to end. It was a good education. But I kept watching the machine learning field move at a pace that felt impossible to track from where I was, and at some point I decided that proximity matters more than comfort.
Why Silicon Valley specifically
I know the criticism. It's expensive, it's insular, the hype is exhausting. All true.
But there's something that's hard to replicate from the outside: the density of people who are building seriously, the ambient level of conversation about what's possible, the culture of treating ambitious ideas as normal rather than eccentric. You can read about it. Being in it is different.
I've been connected to the Draper network for a while. Tim Draper's entrepreneurship program is based here and puts serious founders through a serious curriculum. I'm going to be close to that ecosystem for a while, which means being around early-stage founders, watching what problems they're working on, and having the conversations that don't get written up anywhere.
What I'm actually here for
I have a problem I've been thinking about for a few months. Not ready to talk about it publicly yet, but it sits at the intersection of machine learning and something that hasn't been done well. The research phase is ongoing.
Being here accelerates that. The gap between "machine learning is a research discipline" and "machine learning is something you use to build products" is closing and I want to be on the right side of it when it does.
TensorFlow hit 1.0 in February. Keras made the stack accessible to people who aren't researchers. The moment where a solo founder can build something serious with ML is closer than it looks from the outside.
I'm here to be part of that.
With gusto, Fatih.