> A lot of the value I personally get out of chat-driven programming is I reach a point in the day when I know what needs to be written, I can describe it, but I don’t have the energy to create a new file, start typing, then start looking up the libraries I need... LLMs perform that service for me in programming. They give me a first draft, with some good ideas, with several of the dependencies I need, and often some mistakes. Often, I find fixing those mistakes is a lot easier than starting from scratch.
This to me is the biggest advantage of LLMs. They dramatically reduce the activation energy of doing something you are unfamiliar with. Much in the way that you're a lot more likely to try kitesurfing if you are at the beach standing next to a kitesurfing instructor.
While LLMs may not yet have human-level depth, it's clear that they already have vastly superhuman breadth. You can argue about the current level of expertise (does it have undergrad knowledge in every field? PhD level knowledge in every field?) but you can't argue about the breadth of fields, nor that the level of expertise improves every year.
My guess is that the programmers who find LLMs useful are people who do a lot of different kinds of programming every week (and thus are constantly going from incompetent to competent in things that other people already know), rather than domain experts who do the same kind of narrow and specialized work every day.
I think your biggest takeaway should be that they person writing the blog post is extremely well-known versed in programming and has labored over code for hours, along with writing tests, debugging, etc. He knows what he would like because it's second nature. He was able to get the best from the LLM because his vision of what the code should look like helped craft a solid prompt.
Newer people into programming might not have as good of a time because they may skip actually learning something fundamentals and rely on LLMs as a crutch. Nothing wrong with that, I suppose, but there might be at some point when everything goes up in smoke and the LLM is out of answers.
My experience is opposite - I get the most value out of LLMs for topics that I have less expertise in. It’s become vastly easier up to speed in a new field because you can immediately answer basic questions, have the holes in your understanding pointed out, and be directed to the concepts you are missing.
This to me is the biggest advantage of LLMs. They dramatically reduce the activation energy of doing something you are unfamiliar with. Much in the way that you're a lot more likely to try kitesurfing if you are at the beach standing next to a kitesurfing instructor.
While LLMs may not yet have human-level depth, it's clear that they already have vastly superhuman breadth. You can argue about the current level of expertise (does it have undergrad knowledge in every field? PhD level knowledge in every field?) but you can't argue about the breadth of fields, nor that the level of expertise improves every year.
My guess is that the programmers who find LLMs useful are people who do a lot of different kinds of programming every week (and thus are constantly going from incompetent to competent in things that other people already know), rather than domain experts who do the same kind of narrow and specialized work every day.