What to build with AI
There’s a lot of discussion right now about taste (a good 20 years before ChatGPT got announced) and agency in the age of AI. How do you develop taste. How do you maintain agency when the tools can do so much. These are real questions. But I think they skip over something more basic: do […]
There’s a default assumption baked into most AI product work: the model runs at request time. The user does something, the system calls a model, the model returns a result. It’s how we build chat features, search ranking, content generation — anything with AI in the loop. But there’s a large class of AI-powered features […]
How AI Is Fundamentally Rewriting the Rules of Search For two decades, building a search engine meant roughly the same thing: ingest content, build an index, rank results by relevance, and serve ten blue links. The playbook was well-understood — improve precision, reduce latency, expand coverage, repeat. That playbook is now obsolete. AI hasn’t just […]
There’s a question I find myself asking more often than “should we use AI for this?” It’s the opposite: should we not? The pressure to add AI to everything is real. Every roadmap review, someone suggests an AI-powered version of a feature that already works fine. Every competitor ships an AI something, and the instinct […]
Traditional product specs are built on certainty. When the user clicks this button, this thing happens. Every time, the same way, deterministically. The PM defines the behavior, engineering implements it, QA verifies it. Clean. AI features break this model. The output isn’t deterministic — it’s probabilistic. The same input might produce different outputs. And the […]
Every AI feature starts the same way. Someone builds a demo, it looks magical, and the room gets excited. Three months later, the team is drowning in edge cases and the feature still isn’t shipped. I’ve seen this pattern enough times to know it’s not an execution failure — it’s an estimation failure. The gap […]
Just as you can’t let your report act without understanding the team’s choices, you can’t let AI do the same. Ultimately, you need to understand both what problem is being solved and how—that’s what lets you contribute meaningfully.
A different kind of streak Before Miles, our son, was born I didn’t have a strong opinion about wanting to have kids. Not that I was against it but I didn’t see a strong reason to voluntarily give up a life I dearly enjoyed for an unknown quantity. This feels preposterous in retrospect. But so […]
For those who’ve explored Buddhism, how many times have you heard that Buddhism is about being in the present moment? What often doesn’t get talked about is Buddhism’s relative priorities. Being present is an important concept of Buddhism but it’s not more important than the notion of clinging. Described differently, attachment. We all tend to […]
We just spent a week in paradise and wanted to share some of our favorite spots. If we didn’t love a place, it didn’t make the list. Obviously these recommendations are heavily skewed towards what my wife and I like to do which mostly revolves around being outside as much as possible, eating good food […]
