Here's something most people don't know: you should probably delete your skills and your AGENTS.md every couple of months. The models get better, so how do you know you're not de-tuning them? Unless you've invested in evals over all your traces (lab behavior, and a tiny minority of companies), you don't know whether you're going forward or backward.

That brings me back to the momentum Pai describes. First principles thinking makes working that way feel natural. When you truly understand what you’re trying to accomplish, it’s easier to take a small step, learn from it, and keep going. If you’re doing it right, working with agents lets that back-and-forth happen much faster. You get faster learning loops and more momentum, oriented around deep understanding. To me, that’s the new flow state.

There will probably come a day when I always get sensible answers to these questions that convince me the model knows what it’s doing. But today is not that day. About half the questions I ask get answers that convince me3 the model has made a mistake

In fact, with Astra and Fable I feel like not only are the costs astronomical, but the models are also just not for me as a software engineer. And presumably that’s because these models increasingly are for other people. For lawyers, 3D artists, mathematicians, whoever uses computer use, etc.

I use pi both at home and at work. I'm a bit more daring at home than work. This is something I might try at work.