You sure?

They are targeting the next generation, the next generation will not know how to search the internet without an AI chatbot

We all know that most parents will just let the “Digital Natives” do their thing or/and that kids in a certain age definitely know better than their parents who fear this “technological leap” and don’t get it anyway

So far, there is serious cognitive step needed that LLM just can't do to get productive. They can output code but they don't understand what's going on. They don't grasp architecture. Large projects don't fit on their token window.

There's a remarkably effective solution for this, that helps both humans and models alike - write documentation.

It's actually kind of funny how the LLM wave has sparked a renaissance of high-quality documentation. Who would have thought?

They don't need the entire project to fit in their token windows. There are ways to make them work effectively in large projects. It takes some learning and effort, but I see it regularly in multiple large, complex monorepos.

I still feel somewhat new-ish to using LLMs for code (I was kinda forced to start learning), but when I first jumped into a big codebase with AI configs/docs from people who have been using LLMs for a while, I was kinda shocked. The LLM worked far better than I had ever experienced.

It actually takes a bit of skill to set up a decent workflow/configuration for these things. If you just jump into a big repo that doesn't have configs/docs/optimizations for LLMs, or you haven't figured out a decent workflow, then they'll be underwhelming and significantly less productive.

(I know I'll get downvoted just for describing my experience and observations here, but I don't care. I miss the pre-LLM days very much, but they're gone, whether we like it or not.)

Oh, sorry, I didn't mean to imply that consumer-grade hardware has gotten more efficient. I wouldn't really know about that, but I assume most of the focus is on data centers.

Those were two separate thoughts:

1. Models are getting better, and tooling built around them are getting better, so hopefully we can get to a point where small models (capable of running on consumer-grade hardware) become much more useful.
2. Some modern data center GPUs and TPUs compute more per watt-hour than previous generations.

Yeah true. I'm assuming (and hoping) that the problems with consumer grade hardware being less accessible will be temporary.

I have wristwatches with significantly higher CPU, memory, and storage specs than my first few computers, while consuming significantly less energy. I think the current state of LLMs is pretty rough but will continue to improve.

Man... of all the vibe coding tools, Lovable has gotta be one of the most useless, too.

I work with people (all middle managers) who love Loveable because they can type a two sentence description of an app and it will immediately vomit something into existence. But the code it generates is an absolute disaster and the UIs it designs (which is supposed to be its main draw) is some of the most generic crap I've ever seen.

0/10, do not recommend.

well to be fair you don't even need to look at the md since right at the top it says it's built with loveable.

Who actually tried this?

Got a job application this with a one line cover letter "Iam interested to work with u are company" it was kinda refreshing to see that instead of a whole page of slop, like most of them are these days.

there's at least one guy i know of on github whose claim to fame is he finds code in existing node codebases by big corpos that's duplicated, breaks it out into a library, then PRs the original codebase with "instead of doing manually, switch to depending on this library", then adds to his profile "my code is used by ". he had thousands of libraries like that last i checked, most of them less than ten lines of code. the manifest and other boilerplate is way larger than the actual code.

Your node_modules directory can get so bloated that the community came up with different package managers just for deduplication! pnpm, for example, makes one global-adjacent cache, and then just symlinks the dependencies as needed. This is because the regular npm doesn't, because what if the package changed between the 20ms since I downloaded it for nuxt? (Sorry Nuxt users, had to pick a name)

The worst dependency hell is when a library has a strict version dependency, and another library uses that same dependency. When the second library updates their minimum version of the dependency to one that is higher than the exact version needed for the first, THAT'S dependency hell.