All vibe coding tools are selling a get rich quick scheme

This is the problem. If you couldn't have coded it slowly in the old world, you will have problems coding it in AI world.

However if you have a lot of coding experience, you can now compress the time it would have taken you be an enormous amount. My experience is that I can now make extensive changes with very little effort, and very few dead ends. I've been able to take on entire secondary projects where I was just replication existing knowledge with slightly different tools.

Just this week I had a litmus test. I had an existing database that I'm pushing huge amount of data to. I decided to try a different underlying database. This would have taken me a full week of looking at documentation and writing supporting scripts, now I've done it in the spare time I had in two days of my actual work.

And it's not like the AI just did it all unsupervised. It threatened to do down the wrong path a few times, but each time I spotted it and steered it the way I wanted. I also asked it a few questions about curiosities I discovered in the emitted code, and that led to fixes as well.

If I didn't know how to code before, I would still be coding this alternative database.

Eventually all the experienced people will be dead or retired.

1) that the tools become more Socratic and interactive and educational and walk the engineer through what they're doing

2) juniors pair with a senior who is using the tool and see the process and the decisions being made.

I know the industry wants these things to replace us, but in fact it's more like a power drill than a spinning jenny. It augments and lets the existing craftsmen work better faster, but does not replace / automate really.

I could still see a select few being given a foot in the door and leaping over the "argh why doesn't it compile" stage, past the "argh, I need a whole new architecture" stage to become senior in a few years. Every cohort has a few unicorns.

Alternatively, coding goes the way of the calculator. By that I mean, the people that used to be part of the engineering team, doing arithmetic. You just hire domain experts and give them the AI hoping the domain expertise is all you need to get you through the day.

Finance has historically had a lot of people who thought they could code, now perhaps they will get better. Big risk of it becoming a mess larger than back when they got VBA.

That’s about enough for me to build out fairly complex products very fast with AI (Claude Code). Claude Code often makes some fundamental mistakes that you wouldn’t be able to catch if you didn’t have any coding experience (like today, it was trying to save large images directly in the database instead of using file storage).

Now a junior can ask an AI like 90% of questions that would otherwise occupy a senior:

- Why does this dockerfile copy these files first? - how can I find the entrypoints to this service? - is there anything related to image processing in this entire project?

LLMs let juniors punch above their weight, and it lets seniors go faster. Of course if everybody is twice as efficient, you don’t need as many devs (debatable), but I don’t think junior devs are going anywhere. I hear a lot of CEO hype posts saying junior devs are outdated, but idk it doesn’t make any sense to me!

This is the danger - I get why you think it's true, but it's not true. I've mentored many many bootcamp grads. The biggest danger with y'all is you don't know what you don't know.

Even if you are a really good bootcamp grad and you have good taste and ability, you do not have experience. Software experience is measured really in years and decades, not months.

> You just hire domain experts and give them the AI hoping the domain expertise is all you need to get you through the day.

what domain experts? Clearly they are trying to replace everyone with AI. That's the unsettling part of the whole story; this phenomenon is happening across many sectors who want to try and skimp out on talent.

And how does a Junior in this kind of work gain the skills that transitions them to senior?

>I hear a lot of CEO hype posts saying junior devs are outdated, but idk it doesn’t make any sense to me!

well yes, they aren't making sense. Hence why I think this is all a bubble. Irrational markets and all that.

I don't know when it will burst, but even tech companis can't reject reality forever.

This has not been my experience and it is very frustrating. I've been programming almost 20 years, I'm pretty good at it

I don't know where the disconnect is, but no matter how I try (and I am trying, I don't want to get left behind) I cannot get remotely good results from LLM coding tools

They always, always, always take longer to build what I want than just doing it myself

sadly, "The market can stay irrational longer than you can stay solvent". It will correct itself long term, but the damage over the last few years will linger for years, maybe even decades to come.

Um - by asking AI 90% of the time, and the senior 10% of the time. If anything, the senior can now mentor 10x the number of juniors.

My manager uses cursor and Claude for everything. I can always tell. He pretends like he wrote it himself even, but he obviously didn't. I end up rewriting big chunks of it, or finding all sorts of baffling quirks during review.

Idk, I'm just not at all willing to settle for "good enough," which means Claude doesn't really save me any time. If I'm going to do it, I'm going to do it right.

It's a good rubber duck, and it's good for scaffolding new projects (arguably), and it's good at converting code from one language to another or for generating a one-off bash script or something. That's about all i can get out of it.

"Good enough" often has AWS keys embedded into the code itself, stupid things like that

I hold myself to a higher standard