One of the best use cases for LLMs: they can compile English to code.

The implications of this are significant.

English is at a much higher pace layer than traditional code, which allows moving more quickly and experimenting with much higher leverage.

It used to be that taking an English language description and converting it into code that a computer could actually run was an expensive, tedious, extremely high-skill task of generalization and translation.

But now for software in the small it takes just a few seconds.

As Simon Willison notes, the people best situated for this world are people who both know how to code and also have the knowhow of using LLMs.

A couple of weeks ago I created a little artifact to help with strategy when playing the card game Skyjo.

Writing that little artifact was both very much like programming and also very much not.

I knew how I would have tackled such a project, so I fed the LLM iterative steps, starting with the core of the card counting logic and then adding up features and tweaking them.

But instead of me doing the work, I was just using my knowhow to guide it, and it took a couple of minutes to get precisely the output I wanted instead of hours.

That's the real knowhow of software engineering: What is the simplest possible slice of a thing that will work quickly, that can then be incrementally improved / extended into a production-caliber, full featured thing?