I tried Claude Code for a week straight recently to see what all the hype was about and while it pumped out a bunch of reasonable looking code and features I ended up feeling completely disconnected from my codebase and uncomfortable.

Cancelled the plan I had with them and happily went back to just coding like normal in VSCode with occasional dips into Copilot when a need arose or for rubber ducking and planning. Feels much better as I'm in full control and not trusting the magic black box to get it right or getting fatigue from reading thousands of lines of generated code.

Anyone who says they're able to review thousands of lines effectively that Claude might slop out in a day are lying to themselves.

▲

bluegatty 2 hours ago | parent | next [-]

I don't like calling a posture 'ignorant' , but I think that's what we have here. I don't mean that as an insult.

It's likely you didn't learn how to use the tool properly, and I'd suggest 'trying again' because not using AI soon will be tantamount to digging holes with shovels instead of using construction equipment. Yes, we still need our 'core skill's but, we're not going to be able to live without the leverage of AI.

Yes - AI can generate slop, and probably too many Engineers do that.

Yes - you can 'feel a loss of control' but that's where you have to find your comfort zone.

It's generally a bad idea to produce 'huge amounts of code' - unless it's perfectly consistent with a design, and he architecture is derived from well-known conventions.

Start by using it as an 'assistant' aka research, fill in all the extra bits, and get your testing going.

You'll probably want to guide the architecture, and at least keep an eye on the test code.

Then it's a matter of how much further 'up' you can go,

There are few situations in which we should be 'accepting' large amounts of code, but some of it can be reviewed quickly.

The AI, already now in 2026 can write better code than you at the algorithmic level - it will be tight, clean, 'by the book' and far lesss likley to have erros.

It fails at the architectural and modular level still, that will probably change.

The AI 'makes a clean cut' in the wood, tighter to the line than any carpenter could - like a power tool.

A carpenter that does not use power tools is an 'artisnal craft person' , not really building functional things.

This is the era of motor cars, there is really no option - I don't say that because I'm pro or anti anything, AI is often way over-hyped - that's something else entirely.

It's like the web / cloud etc. it's just 'imminent'.

So try again, experiment, stay open minded.

▲

sassymuffinz an hour ago | parent [-]

Like I said I still use Copilot as needed, I just don't trust Claude to go off on its own and generate a mountain of technical debt that I can't 100% trust.

To use your own analogy, there's plenty of carpenters still around for when someone needs something doing properly and bespoke, even though we can all go to Ikea, or any other flat pack furniture company, to get wobbly furniture cheaply at any time.

I'd rather be the last carpenter charging a liveable wage, working on interesting problems for clients who appreciate a human touch than just pumping out mountains of slop to keep up with the broligarchy. If that makes me ignorant that's fine, but I'll be happily enjoying the craft while you're worrying about your metrics.

▲

bluegatty an hour ago | parent | next [-]

You're offering to deliver parcels by horse - thinking that somehow your 'delivering is better because it's more natural' and that your customers will appreciate it, over the 'smog' that the cars create.

Or in other words - 'non existent'.

It is arrogant and luddite to suggest that 'using AI is not doing it properly' or that anyone will care.

They care that it's done well - that's it.

FYI, the code that AI produces is probably better than what you produce - at least a functional level.

'Artisanility' is worthless in 'code' - there are no 'winding staircases' for us to custom build, as a master carpenter would.

Where you can continue to 'write code by hand' is for very arcane, things, but even then you're still going to have to use AI for a lot of things in support of that.

So if you want to get into compiler design - sure.

But still - without mastery of AI, you'll be left behind.

At least with horses, there's a naturalist component, with 'code' - nobody cares at all. There's zero interest in it, there's not 'organic' angle to sell.

	▲	sassymuffinz 2 minutes ago \| parent \| next [-]
		Maybe in your industry but in mine working with small and medium businesses they value reliability above everything else. They don't give a shit whether you use AI or not as long as it's stable and works and are prepared to pay a premium for someone who knows what they're doing. If you want to have a race to the bottom and be Sam Altman's lap dog, that your business.
	▲	eudamoniac 8 minutes ago \| parent \| prev [-]
		I really don't understand why you people always say these things so matter of factly. I'd put a lot of money (and do, in the markets) on you being wrong. I'm pretty sure in ten years I will not have a problem keeping a software job without using AI.

▲

bitwize 38 minutes ago | parent | prev [-]

No one is paying a liveable wage for purely human-authored code anymore. This is the job now, and you are far more effective with these tools than without. If you still have an issue with their output, that's a PEBKAC and you need to upskill and/or attitude adjust. Stop thinking like a programmer and start thinking like a business person. Delegate! It doesn't matter if the machine wrote code just the way you would have, only that it gets you closer to the goal, and the machine can help with vetting and assuring that it does. If you choose to remain stubborn and closed-minded, what you will find is that clients will not care about the "human touch" in their code, and some AI-assisted consultant will come along and deliver more for less money, drinking your entire fucking milkshake.

In 2005, Tim Bryce wrote that programmers were by and large a lazy, discipline-averse lot who are of average intelligence at best but get very precious about their "craft", not realizing that it's only a small part of a greater whole and it's the business people who drive actual value in a company. AI is proving him 100% correct.

▲

torben-friis 3 hours ago | parent | prev | next [-]

>Anyone who says they're able to review thousands of lines effectively that Claude might slop out in a day are lying to themselves.

The amount you can review before burning out is now the reasonable limit, for the same reason that a car is supposed to stay at the speed you can handle and not the max speed of the engine.

Of course, many people are secretly skipping reviews and some dare to publicly advocate for getting rid of them entirely.

	▲	ethbr1 an hour ago \| parent \| next [-]
		> The amount you can review before burning out is now the reasonable limit I realized this is the crux of our moment, because a variant of Amdahl's law applies to AI code gen. {time gained} = {time saved via gen AI} - {time spent in human review} There's no way that results in a positive number with 100% human review coverage, which means that human review coverage is headed to < 100% (ideally as low as possible).
	▲	sassymuffinz 2 hours ago \| parent \| prev [-]
		> For the same reason that a car is supposed to stay at the speed you can handle and not the max speed of the engine. As we know with driving, sensible drivers stick to the speed limit most of the time, but there's a good percentage of knuckle draggers who just love speeding, some people get drunk, some they just drive the wrong way down the highway entirely. Either way it's usually the sensible people who end up suffering.

▲

coreyburnsdev 3 hours ago | parent | prev | next [-]

why not just use it to review your codebase/commits/prs? you don't have to let it write a bunch of code for you neccessarily.

	▲	sassymuffinz 2 hours ago \| parent [-]
		That's my point - it's great as a tool to talk something through or rubber duck it, but as soon as you just let it loose to slop out thousands of lines a day and never read them all you're really doing is filling your base with thousands of lines of technical debt.

▲

bossyTeacher 2 hours ago | parent | prev [-]

I think most people doing what is now called agentic development, aren't following most established dev methodologies and are to a great extent playing it by vibe.

The codebase disconnect is real.

We are like blue collar workers that need to hit the gym to maintain the body that our cavemen ancestors could maintain by doing their daily duties.

Codebase gym sessions might become a thing.