Having to go through 3rd party services (and rely on them to stay on top of things) to participate is not fun.

I just got back to working on a project I like where we don't use any LLMs for code generation because we think it kind of sucks and leads to bad results unless it's for incredibly small contexts (which is usually not worth the prompting).

You're also not as alone as you think in actually liking programming enough to do it by hand even after having used LLMs. There's also the fact that you build more theory up by actually writing the code anyway, so someone who actually wants to know their code base intimitely will always have to write the code.

I agree that the mismatch between the expectations of OOP as a paradigm and this situation right now feel very similar. I suppose the many billions pumped into selling vibecoding and large-scale code generation via LLMs of other kinds is sort of like the massive amount of money pumped into marketing Java as a viable alternative long before it actually was.

For analysis, bug hunting, overview and some porting work to popular languages and so on I think the current SotA is fantastic, but they're still very disappointing for code generation past function or small module level.

On top of that: Creating software that isn't outright wasteful in terms of performance isn't even hard, it's just a matter of not doing ridiculously dumb things. The problem I've observed in teams I've worked with is that the majority of programmers don't even know what the dumb things are, and wouldn't know how to even approach making something that's halfway fast.

Edit:

Unfortunately I think posts like these are only going to make the problem worse, because now people are going to ask for voodoo solutions to performance issues, when the answer to their problems was usually just "Maybe stop creating wasteful intermediate structures and just walk an array like a sane person" in 99% of cases. The first leg of any optimization journey in the average programmer's code will likely net tens or hundreds of times faster code, and that's actually all people were asking for.

The knowledge required to make those changes and understand them is fairly minimal, but the kinds of people who have to create spinners for webmail interfaces, have their application add 150ms on top of whatever round-trip you have for processing things counted in 5 digits, etc., have never bothered to even learn those things.

This also doesn't even get to the more important point: If you don't have the contents of the phone you have literally no evidence of a crime being committed, other than the one they invented post-facto: "Deleting data that could hypothetically be incriminating, not in any specific way but just generally, maybe".

This isn't an isolated incident, really, I find myself always having these issues with the Gemma series. I'm sure they can do useful things for someone else, but for the things I want to use LLMs for (very small code generation, quick questions, code review) they always seem to disappoint me. I'm sure it's because of the stuff that I do and use, but it's a very consistent red thread with these models for me.

Edit:

The same question for Qwen3.6-35B-A3B produces a pretty concise and correct answer that would be useful to the questioner, without even going to the web. I don't know what Gemma models are trained on, but it's not the stuff that's relevant to me.

It's hard to see why basically anyone would at this point.

Sure, some people legitimately work on things that middle schoolers can put together, and for those things I suppose using LLMs as a way of just getting them done is a less risky prospect. For those cases I don't really see the point of tossing it over the hedge to LLMs because what exactly are you doing beyond that? These types of projects have almost nothing to actually do beyond the grunt work of wiring together MVC stuff.

It's not as if you're going to now have time for interesting things, as they tend to not exist at all in those types of projects, and they're rarely important or big enough to warrant much thought in terms of distributed systems design and so on.

I guess I can see a situation where someone instead finds time to work on actually interesting side projects because they've basically stopped doing their work manually because it's not something worth doing manually. Personally I don't think I could stand that type of situation because it would feel bad on a moral level as well as on a personal stimulation level.

Or better yet, just don't use agents, deliver something better with all the stability and boringness that you have. There's never been a better time than now to produce software that actually seems to work and march along steadily; your competitors are all flailing wildly like morons.

It's not true as a general rule that they could give you correct small code snippets five years ago; I had frontier models not even understanding enough of the OpenGL API to make separate framebuffers correctly. I asked this as a test and the resulting recommended code would of course compile, but leave someone who didn't already know the correct way to do it staring at a black screen and potentially spending the next hour debugging that.

If you think you're somehow going to make an actual game worthy of people's time with LLMs, you are almost certainly mistaken. Not only because it'll be a complete, unworkable mess with really bad fundamentals, but also because if you're vibecoding the actual technical part you probably also think game design is just about "having ideas" or something else that's really reductive and incorrect.