Most of my time goes to two things: pointing a team at the right problem, and then getting out of its way. The work I'm proudest of isn't a screen — it's a group of people who stopped waiting on my decisions and started making better ones than I would have.
Setting direction is the first half: the first outcome-oriented OKRs at Flatpay, KPIs and customer signal close enough to hand that teams can decide for themselves. It's moved real numbers — product NPS from 25 to 55, four-fold ARR growth, a product org that grew from three people to more than twenty. But the numbers are downstream of the structure. Durable, empowered teams produce them; heroics don't.
AI is part of how I work now, not a line in the deck. I keep a model open the way I used to keep a terminal open — for research, for turning customer signal into something usable, for building the thing instead of describing it. I shipped four apps this year mostly to find out where it genuinely helps and where it quietly doesn't, then brought that back to the team.
