A year ago, cursor was flummoxed by simple things Claude code navigates with ease. But there are still corner cases where it hallucinates on the strangest seemingly obvious things. I'm working on getting it to write code to make what's going on in front of its face more visible to it currently.
I guess it's a question of where you find joy in life. I find no joy in frameworks and APIs. I find it entirely in doing the impossible out of sample things for which these agents are not competitive yet.
I will even say IMO AI coding agents are the coolest thing I've seen since I saw the first cut of cuda 20 years ago. And I expect the same level of belligerence and resistance to it that I saw deployed against cuda. People hate change by and large.
Lots of people wanted (and Intel tried to sell, somewhat succesfully) something they could just plug-and-play and just run the parallel implementations they'd already written for supercomputes using x86. It seemed easier. Why invest all of this effort into CUDA when Intel are going to come and make your current code work just as fast as this strange CUDA stuff in a year or two.
Deep learning is quite different from the earlier uses of CUDA. Those use cases were often massive, often old, FORTRAN programs where to get things running well you had to write many separate kernels targeting each bit. And it all had to be on there to avoid expensive copies between GPU and CPU, and early CUDA was a lot less programmable than it is now, with huge performance penalties for relatively small "mistakes". Also many of your key contributers are scientists rather than profressional programmers who see programming as getting in the way of doing what they acutally want to do. They don't want to spend time completely rewriting their applications and optimizing CUDA kernels, they want to keep on with their incremental modifications to existing codebases.
Then deep learning came along and researchers were already using frameworks (Lua Torch, Caffe, Theano). The framework authors only had to support the few operations required to get Convnets working very fast on GPUs, and it was minimal effort for researchers to run. It grew a lot from there, but going from "nothing" to "most people can run their Convnet research" on GPUs was much eaiser for these frameworks than it was for any large traditional HPC scientific application.
It seems funny though: The advantages of GPGPU are so obvious and unambiguous compared to AI. But then again, with every new technology you probably also had management pushing to use technology_a for <enter something inappropriate for technology_a>.
Like in a few decades when the way we work with AI has matured and become completely normal it might be hard to imagine why people nowadays questioned its use. But they won't know about the million stupid uses of AI we're confronted with every day :)
Won't name names anymore, it really doesn't matter. But I feel the same way about people still characterizing LLMs as stochastic parrots and glorified autocomplete as I feel about certain CPU luminaries (won't name names) continuing to state that GPUs are bad because they were designed for gaming. Neither sorts are keeping up with how fast things change.