Comment by zmmmmm
6 days ago
I think in a lot of cases the skeptics haven't really witnessed how good LLMs can be for a subset of software engineering tasks. I find for mid-complexity code they are far, far better than I am at coding. Even though I often tweak their solutions, they outweigh that by often identifying APIs and even whole libraries to call where I fully would have set about implementing it myself - either out of ignorance or an assumption that it would be too hard to implement it myself. They also put in much more error checking than I would normally do. Even if I do implement something myself, it's insane not to at least see what the LLM would do to get ideas. The thing is, once I do that, I nearly always come away with the conclusion that I might as well just use the code it made.
The primary skill a developer hones is the ability to know when something is already solved for you. This is where innate laziness is a gift. Why should I write a fetch function if api.fetch exists? This is a natural game time automatic decision, like a baseball player running for first base soon as the ball hits the bat. The natural thing that should be happening to every developer is the natural realization that the LLM is the ultimate api. I have literally sat here going “well why build anything, this damn thing does this and that for free”. For example, why build a todo app? Any of the LLMs can mostly do this. This is not self awareness, but game conditioning and fitness. You should automatically be feeling the sensation just like someone playing a sport for the millionth time.
Developers thinking AI is a fad is like a left fielder thinking about if they should stick their hand out to catch the ball or not. You can’t think about this.
Anyway, I find myself in stasis trying to actually see what to build given the reality of the power of this api. The South Park episode of “The Simpsons did it” is an apt episode. We are very much at the point where the continuous question will be “Well the LLM can do it”.