Just a couple days ago, I attended a python meetup where an experienced developer with a couple years experience using LLM assist tools, estimated that they gave him 10x the productivity.
The main thing to take away from studies like this, is not whether or not LLM-based tools are useful (clearly they at least sometimes are). It's that it cannot be the case that the breathless estimates (like "10x") are correct, or even a flawed study (as long as it was not fraudulent) would find some improvement.
If you had done this kind of study when we first started using higher level languages instead of assembler, we would not have had to worry about whether or not we made some subtle (and common) statistical error. LLM tools clearly are useful, at least sometimes, but we haven't seen Microsoft (early adopter of such tools and early investor in OpenAI) have some big burst of productivity. So, they cannot be the unprecedented game-changer that their boosters allege.
What I would like to see is some kind of study of how well it helps new developers, or experienced developers using a new language or programming in a new problem space. That seems like the kind of thing that would get the biggest improvement from LLM tools.
cranberryturkey · 4h ago
I have been coding for 30 years now. I can do in 20 minutes or an hour what used to take 2 weeks.
The main thing to take away from studies like this, is not whether or not LLM-based tools are useful (clearly they at least sometimes are). It's that it cannot be the case that the breathless estimates (like "10x") are correct, or even a flawed study (as long as it was not fraudulent) would find some improvement.
If you had done this kind of study when we first started using higher level languages instead of assembler, we would not have had to worry about whether or not we made some subtle (and common) statistical error. LLM tools clearly are useful, at least sometimes, but we haven't seen Microsoft (early adopter of such tools and early investor in OpenAI) have some big burst of productivity. So, they cannot be the unprecedented game-changer that their boosters allege.
What I would like to see is some kind of study of how well it helps new developers, or experienced developers using a new language or programming in a new problem space. That seems like the kind of thing that would get the biggest improvement from LLM tools.