Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The last six months in LLMs in five minutes

simonwillison.net

Post reply on HN
Print view View original

541–550 of 631 posts

Re: The last six months in LLMs in five minutes

#541

Post by magicalhippo »

skydhash wrote: post #417
magicalhippo wrote: post #251

Earlier quoted context omitted.

For me, the fun in programming is sometimes to actually write code, solving a problem in a specific way or try some new approach. Other times the fun is to create something that works, and the code is more a means to an end. The first case I'll probably still do by hand, like handmade vases despite factory made are cheap and readily available. For the second case I think these newfangled tools have made it even more…

> I think these newfangled tools have made it even more fun, since writing lots of boiler plate, repetitive event handles and whatnot is not my idea of fun That’s what code generators, snippets plugins, macros, and the old copy-paste are here for. I wonder if you were using notepad to code. Because even nano had macros.

Those tools only get you so far, especially if you write something novel to you. Using a new framework or programming language say.

Sometimes using a new framework or programming language is the fun part.

But sometimes it's just the best way of solving a problem incidental to the fun part.

One of the two projects I vibed included a web frontend. I didn't touch a single line of HTML, CSS or JavaScript of the frontend. And I didn't touch the API on the backend. I'm not a web dev, so this isn't something I've got snippets for or whatever, and in this case wasn't the interesting part.

The interesting part for me in that case was making a tool that could help us, not the details how exactly how that was done.

Re: The last six months in LLMs in five minutes

#542

Post by mekael »

dawnerd wrote: post #181
OakNinja wrote: post #123

Earlier quoted context omitted.

To me, LLM's free up time for me so that I can spend time on the fun parts of coding. Less boilerplate, more focus on the interesting problems. This is no different from using high level languages. The problem domain is less around memory management and garbage collection and closer to the problem you're actually trying to solve.

But we’ve had tools to automate out the boilerplate for years. We don’t need ai for that. It’s seriously like we all forgot we could run one command and scaffold a project. AI isn’t even that great at it. Last I tried a month ago it used a really out of date version of nextjs and picked all sorts of random deps that weren’t in the plan. I could have just used the next project scaffold tool and been on my way before t…

You have no idea how many times I’ve asked “why are we not using the project generator” or “why did you write 200 lines to parse a csv? Here’s a library and five lines to get it done” in the last year. Its easily up 20x compared to pre ai, and getting worse.

Re: The last six months in LLMs in five minutes

#543

Post by kstenerud »

wasabi991011 wrote: post #520
kstenerud wrote: post #371

Earlier quoted context omitted.

It's more along the lines of: Anyone with an axe to grind will find something to grind it on. Early stage products will have some rough edges. We've seen that in Docker, Kubernetes, AWS, Azure, LXC, KVM, etc. And people griped and raged about the sheer incompetence of the maintainers and utter lack of quality, but they still used those tools even before the rough edges were polished away and folks finally settled dow…

> Early stage products will have some rough edges. We've seen that in Docker, Kubernetes, AWS, Azure, LXC, KVM, etc. Is your product really the same complexity as these?

It tackles similar kinds of problems, dealing with idiosyncrasies in Linux distros (and Mac), docker, containers, kata, firecracker, seatbelt, tart, tmux, Claude, the various terminal emulators out there, and trying to herd those cats such that it doesn't blow up in your face.

Is it doing it to the same scale? No - it's a single user app. But have a look at https://github.com/kstenerud/yoloai/blob/main/docs/dev/backe... and you'll see the kind of shit a project like this has to handle. It's not trivial.

Re: The last six months in LLMs in five minutes

#544

Post by bradley13 »

opto wrote: post #535
hug wrote: post #448

Earlier quoted context omitted.

I don't mean to tar you with a too-wide brush, and I feel like you have a good handle on your personal acceptance for LLM assistance. No complaint there. I do think, maybe alternative to your view, that LLMs can provide useful feedback to graduate-level employees in most fields. It is not that the work can be done by LLMs -- we're not there, yet, in software or otherwise -- but that LLMs as useful tutors specifically…

What does that really mean though — ten more years of data centers exploiting local communities for their resources will mean that a computer might be able to teach people to tie knots, and reliably check their work... No government would allow that to certify someone, and no company would risk the lawsuit when someone dies doing what the AI tells them, so it's a non-starter. Even if it were possible, and governments…

> "What does that really mean though — ten more years of data centers exploiting local communities for their resources"

That is purest hyperbole. Data centers use a lot of electricity, but they are hardly looting local communities. The water issue is wildly exaggerated, unless a data center is located in a desert, because most water is recirculated.

And why do you think no one will allow an AI to certify someone on certain topics. Their knowledge at the moment is roughly the average of people in the field. Is an average person in your field not able to certify others? In any case, AIs are improving very rapidly, so what is not possible today will be possible tomorrow.

As an example, let me point out the Tesla FSD. On a per-mile basis, self-driving Teslas have a massively lower accident rate (less than 20%) than human-driven vehicles. That is a very physical activity being handled by an AI.

Re: The last six months in LLMs in five minutes

#545

Post by my-next-account »

walthamstow wrote: post #278
Philip-J-Fry wrote: post #248

Earlier quoted context omitted.

I don't want to offend (it's AI coded anyway :)) but that does not scream "high quality" to me. The headline gif on that repo just paints a terrible picture. It can't draw a box correctly, there's random underscores all over the screen. The UI itself is just incredibly incoherent. I don't even know what I'm looking at. Like, no it doesn't seem like very high quality work... It just seems like a vibe coded tool. Edit:…

Take it up with Anthropic. It's actually their billion-dollar TUI product you're commenting on. The problem with being such a naysayer is that you're entirely disconnected from what's going on. You haven't tried an agent like Claude Code and experienced it for yourself, so you don't recognise what it looks like when it's in front of you.

I have used those tools, I don't think they're THAT good tbh :P

Re: The last six months in LLMs in five minutes

#546

Post by topherhunt »

hollowturtle wrote: post #209

> The coding agents got really good It's since november 2025, the so called "inflection point", that I'm still wondering for who coding agents become "really good". All I observe they got better at tool call and answering questions about big codebases, especially if the question has a vague pattern to search, and they're superuseful for that! For generating production code even with a lot of steering and baby sitting…

@hollowturtle I'm surprised - do you really find that sota models aren't good enough to generate production code with steering and babysitting? My experience (Claude Code, mostly Opus 4.6) is that it's fantastic at this. At least in JS + TS + Elixir + Ruby. It does indeed need babysitting, my mental model is that it's an exoskeleton not a junior dev, but IME it's a friggin badass exoskeleton, easily 10x-ing my speed on most work. Notably I do NOT --dangerously-skip-permissions nor use claude code's auto mode, I micromanage and lightly review every line it's writing as it writes it, so I rarely have more than 2 sessions generating simultaneously. I suspect that a lot of the disappointment comes in when people try to delegate to it and trust it to not go off the rails. It hasn't earned that trust from me yet (and hasn't needed to yet).

Granted, I'm mostly working in small-to-medium codebases, 20k-30k LOC incl test suite. I wonder if that's a factor in my positive experience. Curious to hear your thoughts.

Re: The last six months in LLMs in five minutes

#547

Post by max_unbearable »

The honest summary that doesn't show up in the six-month roundup: the unevenness. Boilerplate, tests, scaffolding, glue code: dramatically faster, sometimes 5-10x. Architecture, data modeling, careful security work, judgment calls about what to build: same as before, sometimes slower because tab-completion sneaks in plausible-but-wrong defaults you then have to undo.

The thing headline numbers ("AI made me 3x faster") hide is which 30% of the work the AI sped up and which 70% didn't move. For a solo dev the survivable bet got smaller, and that's the real change, not raw productivity. AI made certain projects worth attempting at all that wouldn't have been viable six months earlier.

Re: The last six months in LLMs in five minutes

#548

Post by 3l3ktr4 »

I'm always surprised to see HN people saying models aren't good. What are these guys building? The best engineers I know, from startup to big tech admit these models are incredible. Including people I don't know personally, foundational engineers from every area. The average HN person though, is doing some quantum-alien computation that not even the best developers in the world can grasp.

Re: The last six months in LLMs in five minutes

#549

Post by liuliu »

randusername wrote: post #528
hollowturtle wrote: post #459

Earlier quoted context omitted.

c/c++, java, kotlin, go, some perl scripting, some javascript. Gaming industry

We'll there's your problem. F1 mechanic pops the hood of a mass-market Toyota Corolla and doesn't understand why everyone says it's really good. A lot of us are out here building websites or phone apps. Not to say that these things can't also be taken very seriously from first-principles, but I think that's rare.

Probably not really. For gaming, I think probably just need to have a better way to explain visual and what the problem is (collision not done correctly, ways to feedback to LLM's experimentation loop how that should be checked and why etc).

Models usually is broken if there is no feedback loop. Well, websites might be exception since they can one-shot pretty well. But there are plenty of things they can do well without one-shot that just requires a good feedback loop to be built.

Re: The last six months in LLMs in five minutes

#550

Post by mwigdahl »

y0eswddl wrote: post #446
hollowturtle wrote: post #228

Earlier quoted context omitted.

Please do not cite Dunning–Kruger effect at random. Who needs to generate a dumb demo of a 97% done crud app? We had code generators for those, everytime I read claims like that and I ask to explain further I then discover it's people who were not productive before generating the so called "MVP level things to completion with ease". If you're trying to solve a HARD problem people REALLY have, it's a novelty that agen…

I'm beginning to get the sense that Sturgeon's Law is at play here and the non-crap 10% of us are arguing with the 90% for whom LLM's shitty output is actually better than what they could do on their own. I've been lucky enough to work at places with majority intelligent engineers with similar tastes on quality to my own... but it seems to be that's not the norm or the case everywhere. and it's the 90% that's most vo…

“Am I out of touch?”

“No, it’s the children who are wrong.”

Post reply on HN