During my computer science studies, our introduction to artificial intelligence didn’t begin with neural networks or robotics, but with a parade of definitions:

• “We call programs intelligent if they exhibit behaviors that would be regarded intelligent if they were exhibited by human beings.” — Herbert Simon

I've been making Fakemon (fake pokemon) using a combination of traditional sketching and modern AI tools. The journey takes us from initial concept sketches through ChatGPT refinement, and finally to Sora for animation.

A Product Manager is supposed to be the great connector—the person who brings engineering, design, and business together. PMs are expected to be curious, collaborative, and customer-obsessed. They align stakeholders, navigate ambiguity, and create clarity from chaos. They're not the boss, but they lead. Not the expert, but the glue.

I recently ran a values exercise with a team of engineers in 1-1s. Each had to select three sticky-notes from a predefined set of values. Every single one of them chose Cooperation.

When we asked what cooperation meant, I got five completely different answers:

As a teenager, I was all ideas and no follow-through. Every week brought a new concept, a new scheme, a new startup in my head. I wore my imagination like a badge. But over time, I had to confront something uncomfortable: ideas weren't rare. They were a ten a penny, a kind of noise that kept me from actually doing the hard work.

The push to remove feelings from product management is misguided. We've somehow convinced ourselves that good product decisions must be purely rational, backed by data, stripped of intuition. This is mad.

I've been doing a lot of vibe coding, and my particular tools of choice include v0.dev (as a Vercelvercel Vercel is a cloud platform for frontend developers. It is used to deploy and host websites and web applications. fanboy). By default, v0 will use tailwind for styling. This means I've been using Tailwind for the first time, and... I hate it.

In a recent piece, I made the case that we should stop trying to build "perfect" AI. That imperfection is not a failure mode — it's intrinsic to how these systems work. Here, I want to go one step further: not just to excuse AI's flaws, but to explore how we can use them. How we can design with imperfection in mind.