I couldn’t agree with this more strongly. It’s only when you know the nooks and crannies of the way your users’ problems actually exist that you realize that first principles thinking is a great place to start your journey but a terrible one to persist on. That’s the kind of

But what does it mean? Everyone talks about how data is key thing but I'm not so sure. There's data everywhere - but making it mean something? That's the hard bit. To make agents work we need to make sure they know what we mean when we say - order bday present or reschedule

Success is not linear:

“Modern luxury is the ability to think clearly, sleep deeply, move slowly, and live quietly in a world designed to prevent all four” — Justin Welsh

“Agents are here, but in order to be useful they need access—access to your computer, to other apps, to other agents, and to the internet.” Agents mean access, and access means trust. We can build all the agents we want but unless people trust them, they'll be a fraction of

Most risk isn't real

You can't live a big life in a small environment.

Improve your attitude

It is much easier to advise 1000 companies than build 1.

Wise words from Socrates:

OK my AI assistant startups - Can we get some Stevens and Garys for every Donna or Peggys we’ll all be barking out orders to? Let’s not bake our futures based on the TV shows of the past.

Nothing makes me more insane than people who have never volunteered at a school, coached a kids’ team or taught a class going on about everything that is broken and how they have the answer if the dummies would get out of the way. Yes lots of things are broken. But you have to

this is gold

Bezos once said: "The thing I've noticed is when the anecdotes and the data disagree, the anecdotes are usually right. There's something wrong with the way you are measuring it." Applies to so much more than business. Never let a statistic blind you from seeing the naked truth

All you gotta know

C. S. Lewis, live near your friends

my new rule - if something *almost* works in an AI product - but doesn’t quite get there, for either cost or model quality / reliability reasons, set a reminder and look at it again in a month chances are it will work now

Actually, this is a recipe for a really rewarding life if you’re a recent college grad not sure what to do: Find a small town you could love with a local newspaper whose owners are ready to retire. Raise the capital to buy it. Run it with the community’s interests at heart.