Capricorn2481

> If it can be measured, then LLMs can optimize it

Then they can start attempting to optimize it. They can also spin round and round making the numbers worse because they don't actually know what to do.

> The personality is bland

I don't use Grok, but do you want your LLM to have a personality? "Personality" is exactly what people don't like about Claude.

> That's why it struck me as being so weird; something so innocuous on its face getting hammered that badly suggests some kind of outside coordinated influence.

That seems like circular logic? Hasn't dang offered many reasons why a post might be suddenly down rated?

There's just got to be more than "I think I saw a post get down rated, so it must be a bot army"

Well I definitely think the site is heavily botted, as it's trivially easy to do so. I was speaking more to moderation.

This isn't very satisfying to me, as that ranks pretty low on the list of potentially damaging threads, don't you think? The genesis period of a petition? I see way more alarming stuff on here every day that further disillusions me to the biggest players.

I don't like Paul Graham, but I don't know if he really gives a shit what's happening on here. HN is not a super influential forum.

I think you're completely right to by skeptical, and I mostly agree with you. I'm also not that familiar with his work.

But to play devil's advocate, working at consulting companies is probably the best way you're going to get exposed to the outcomes of different design decisions. I feel I've learned a lot more being brought onto different legacy projects than I have starting a dozen greenfield projects at the same company, where you may not even stay long enough to see if that trendy design choice pays off. You also get very used to seeing different management/programmer dynamics.

So I don't think working at only two consulting companies is a bad thing.

If you read the comment for more than half a second you'd see they're not questioning why someone would use PHP. They're asking why you would use this particular server over nginx/apache, given how robust their support is.

Signed, a PHP developer.

> Nowadays RAM is expensive because many people now want to max out their computers with RAM to run LLMs and AI coding assistants.

Source? I imagine the amount of people trying to run local LLMs is miniscule. RAM is expensive because a handful of companies have spent billions buying all of the compute.

What a tedious reply. They're genuinely asking what the technical reasons for wanting that would be. You're reading it in the silliest way possible.

Thanks so much for this, I'm using it all the time. I self host a few things, but I'm using this the most.

I think if you could drink alcohol through your phone, people would probably pause on whether that's a good thing.

Surely you can use your imagination a bit and imagine the difference between baseball cards and connecting your bank account to a gambling app, especially with all the paid endorsements from companies and celebrities telling young people they will make a lot of money using these apps.

> 98% of people think AI means what gemini tells them when they do a google search.

> of that 2% who go beyond... maybe 20% of those are using AI to code.

> so of that 20%, maybe 5% have... maybe 1-5% of that 5% actually went deeper

How can you write any of this when you said you just started liking LLMs with Astra, a model that released two weeks ago. You're whipping out a bunch of made up stats trying to describe large swaths of programmers while admittedly being unfamiliar with the capabilities until two weeks ago? Why are you so antagonistic towards people's skepticism if you used them for years and found no value in them? This sounds more like a comment trying to generate massive amounts of FOMO. "If you REALLY put a lot of your money into them, that's when they shine! Don't believe the haters!"

> like i said, if you are still having quality issues in 2026, that's a skill gap.

> also, agents now continually improve the process.

> for a business, the only thing that matters is transaction log. for 99% of businesses, swe are a cost.

Maybe you should try building something for more than two weeks before making such big declarations, because this really does read like someone who is excited about a novel they wrote last night at 3 am

The actual question is backwards: how do we keep the prompt and context small enough so the LLM doesn't start hallucinating basic rules of chess.

Because they are obviously testing for general intelligence. If you want a thread about how cool the harness is, that's down the street.

We don't really consider humans downloading stockfish to beat people at chess as noteworthy endeavors.

Definitely not. A human can write a chess engine without being very good at chess.

I think using AI for customer service is really, really ineffective, and it's plain to anybody that has interacted with it.

It's a shame that we've had decades of shitty customer service from companies that have the most responsibility and resources to do it, and people in tech have shrugged their shoulders saying "it's unreasonable to expect people to scale up their customer service! They gotta make money!" Now AI customer service has arrived and things are worse, not better. Now we can't even get in contact with anyone because an agent redirects us to documentation instead of directing us to a human being.

The future, when a company steals your money, is making viral posts online trying to get enough public shame going that you will maybe get some of that money back.

> I'd like to turn up my internal sense of skepticism when dealing with someone, effectively, selling something their salary depends on

This is really, really silly. Java is many times beyond the position where its developers need to desperately convince people to use it. This is a person who has unique technical expertise in the area whose credentials are smack dab on their profile, not hidden from you. Their closeness to the domain at hand should make you less skeptical of what they are saying.

> Saying that Go lacks in those is just showing how people are making software those days. It’s just terrifying

I'm not following. Saying Go lacks in X shows how poor other software is? Can you connect the dots?

This article is 2 months old when the event happened. And they didn't "bill them", they politely asked them on Twitter, and it looks like that went nowhere.

> Black mirror the show is a good example, also considered intellectual by many

Absolutely not.

I don't think that really matters. I have already seen companies generating their own code and just QA testing it, with no software engineers in sight. Or they ask us to "review" their 10k loc PR in 2 hours.

The spirit of the parent comment is true. People feel like it's productive and even if they make something worse, they are going to use it.

Hard disagree. It adds context to what they said.

These things are borderline useless with web search. It's amazing that they just throw out their entire training data and read you the first three things they found on the Internet.

You wanna try reading the whole comment?

Most developers cannot make any experience on Native.

I would rather have the opposite? I don't want 10 people making small apps that use Electron. I would rather that be used for bigger apps where the overhead is negligible.

Huh? You're right that South Park has had a "both sides bad" approach to politics in the past, but I think you're misrepresenting how the show covered Trump. South Park was pretty much shitting on him the moment he announced his campaign and they painted him as a malevolent force that was going to ruin America, the entire time.

They kinda made fun of Hillary, but nothing beyond she's lame and a bad campaigner. And they were a little hamstrung there considering they were jumping ahead of the real life election cycle a bit.

Like I said, in my experience I would rather do this with a bunch of tiny prompts than try to craft something that oneshots it. Chances are it will do several things I don't want, no matter how clear I think I'm being.

Maybe. Do you have an example of a prompt you've done recently?

I don't think it's valuable at all to try and craft a specific prompt to "one-shot" a task, I'm way faster just doing small asks and guiding it. At 1t/s, how would you even know your prompt was insufficient? It would take you 7 hours to see it.

I don't really do anything at my job that could be done in a single prompt, and certainly not something that would take me 7 hours. A 7 hour task would take multiple iterations with how LLMs are right now.

You said it would do overnight what would take a human half a week, so I'm curious what tasks you are doing where AI is 3x faster than you even at 1t/s