The problem is that the lines of code are riding on a stack of other dependencies that all need care and feeding. Things reach EOL. Frameworks have major breaking changes. CVEs are discovered.
You can actually get high-quality code out of them -- at least with Claude; not had a great experience with Gemini -- but for complex tasks requires riding them very, very hard and really understanding where things can go wrong and poking at them repeatedly. Iterate, iterate, iterate.
That describes my last week. What made it most annoying, was the need to release through TestFlight, because the memory issues would not appear, when tethered. Also, I was checking in constantly, because I had to revert and reset the context, several times.
Yes. It's also why working as a software (host) developer at a hardware company is difficult.
Hardware people insist on treating software the same as firmware.
Bad firmware can cause real-world, physical damage, and be impossible to fix without a hardware recall. A firmware bug can wipe out a hardware company. A software bug can be embarassing, but can also be corrected a lot more easily (as long as it is being treated differently from firmware).
> So far, LLMs seem to deliver code with "Louie Da Loan Shark"-levels of tech debt.
Maybe a couple of years ago, but these days, Opus 4.8 is frankly writing better software than what I've seen over the previous decades in non-tech enterprise. These previous two months, we've replaced so much technical debt we've been dragging along for the previous 5 years as our team went from 25 to 3 people.
This is in non-tech enterprise in Denmark and AI had absolutely no impact on us going from 25 to 3. That was all Putin and bad business decisions on the c-levels. Like keeping flexible loans to fund projects on the books when the interests rates were 0.01% because they might go to 0.001%. Anyway, I'm getting to the point where the AI does 100% of the work, but only if it's piloted by people who know what security, resource consumption and compliance is. The code itself is excellent though.
It likely depends on the implementation, and the tool.
In my work, the Swift code is not really that good (but it’s not terrible), but the PHP code is very good (better than mine). I use ChatGPT. Maybe Claude might give better Swift, but I’ve invested quite a bit of context in ChatGPT.
Your experience is apparently different than mine. I went from using our corporate tool to copilot cowork when it became available to us. From opus 4.6 to 4.8 and there has been a massive difference. It's ridiculously good at programming in the right hands, but the right hands is frankly becoming more and more automatable as well, since you can input design documents, compliance policies and allowed packages and it'll do fine.
If you want, you can go through my history and you'll find that I haven't exactly been a fan of AI, but it's silly to deny that it's gotten good.
Yup. In another post, I was grumping about having to accept a truly obese bunch of code from an LLM. This particular issue is the only one I could imagine even considering accepting that much pasta, but I sort of have to, because the LLM was able to quickly solve a problem that would have taken me a couple more weeks to address.
I remember a .sig that went something along the lines of:
I hate code, and want as little as possible in my software.
You're acting like the landscape of law hasn't changed and been affected drastically by the purchased Supreme Court justices. That has literally happened. It's not a hypothetical or speculation.
If I had a few dollars for every time I heard “that’s so beyond the pale, it couldn’t possibly happen”, I would be buying some more RAM for my homelab right now.
> Right out of the gate nearly all retail investment platforms have dramatically reduced requirements for purchasing an IPO, most notably Fidelity, which previously required $500,000 in your account to participate in an IPO reduced (on Friday) this amount to $2,000
Not at all surprising that the US in 2026 has degenerated to the point of turning the equity market itself into a bucket shop.
> we have the rich investing in existing housing because we have effectively stopped production
In my quite affluent city, the most affluent neighborhood has yard signs up every fifty feet expressing that they are furious that the city might develop city-owned parking lots (which are nowhere near their neighborhood) to be high density housing.
One-bedroom apartments in the city center are going for over $3,000 per month.
The rich aren't simply locking up existing housing, their principal concern is preventing any housing from being created, even if it has no effect on them.
I think there should be a relatively simple solution. Tax the homeowners proportional to their homes' market value. Either you get enough tax revenue to build more houses, or the tax burden is too high for those NIMBY to remain at their beautiful suburban homes, and their houses will be back on the market.
But no, when we propose that, all those affluent bourgeois feeding on the young are suddenly poor grandmas just wanting to live the rest of their lives in peace.
Make up your mind, people. Are we going to fix the problem, or are we going to blame Jeff Bezos because some suburban schmucks oppose building apartments. Frankly I don't think Jeff Bezos even cares about apartments. (He has enough money to buy a city block, why would he care.) But it sure feels more righteous to hate Jeff Bezos than argue about real estate tax.
Most states (and other countries) do tax the homeowners proportional to their homes’ market value. California with its tax proportional to value at purchase is the outlier.
I think it’s largely just that people just don’t want to live near poor people, because they think they have bad culture/values/behavior, and will be risky to live near.
Based on comments I’ve seen at city council meetings about this stuff, there’s also some aspect of feeling like infrastructure is already overstressed, traffic is already bad, etc, which is largely an artifact of car-centric development patterns being incredibly wasteful/inefficient, and capping out at relatively low densities. But the existing development pattern is usually not a good fit for mass transit - the utilization is usually too low.
I think the California approach of aggressively upzoning near public transit is pretty good, except that it might cause resistance to public transit expansion.
This kind of thing becomes a self-fulfilling prophecy when all the poor people get crammed into one place.
Singapore does it right by having high quality housing which happens to have a certain amount subsidized for lower income people. You get a mix of incomes and not a slum.
A lot of California's housing development also incentivizes this type of arrangement: permitting can be fast-tracked and local NIMBYs can be steamrolled if a development allocates some, but not all, of the development to be designated as affordable.
Yeah, no argument. I think many in the US want the bar to be higher, though - many want expensive single family housing to be the minimum in their area, apartments would be too affordable, even unsubsidized.
Every day another form of the Onion's Stock Market Invincible - "Buy buy buy! Experts advise" headline.
SpaceX's whole IPO is about them really being an AI play now (with some minor fun side hustles), except their AI strategy is just them buying and leasing a lot of depreciating nVidia hardware. The hardware is getting more expensive, the costs of operating it will increase, and the value of the tokens will drop.
All this in an environment where voters are freakishly united on loathing of new data center buildout.
All the market news today is because a strong jobs report dropped, which shuts down any notion of rates going down. This of course seriously hampers casino operations.
Pretty sure some other principled people stayed behind and are using the field manual[1].
[1] https://gutenberg.org/ebooks/26184
reply