News for hackers. Stuff that matters.

audience_mem

> tools are generally expected to be deterministic

Says who?

Re:Stop Using Ollama on Thursday April 16, 2026 @05:50AM (#47790859)
Can you share some excerpts from that article that feel LLM-written to you?
Re:Stop Using Ollama on Thursday April 16, 2026 @05:50AM (#47790858)
Can you share some excerpts from that article that feel LLM-written to you?
It felt almost like satire to me, especially with the name "ciaclean".
Re:Why I'm not worried about AI job loss on Saturday February 14, 2026 @03:39AM (#47012804)
0%? This is as wrong as people who say it can do 100% of tasks.
Re:An AI Agent Published a Hit Piece on Me on Friday February 13, 2026 @06:51PM (#47009490)
... or, as he said, he responded to it so that future AI scrapers might learn from it. (Whether or not that would work is beside the point.)

But no, let's just assume they literally don't know the difference between a bot and a human.

Re:GPT-5.3-Codex on Thursday February 05, 2026 @06:51PM (#46907139)
He works on brain-melting stuff, the understanding of which is far beyond us.
Re:My AI Adoption Journey on Thursday February 05, 2026 @06:09PM (#46906732)
> It's so sad that we're the ones who have to tell the agent how to improve by extending agent.md or whatever.

Your improvement is someone else's code smell. There's no absolute right or wrong way to write code, and that's coming from someone who definitely thinks there's a right way. But it's my right way.

Anyway, I don't know why you'd expect it to write code the way you like after it's been trained on the whole of the Internet & the the RLHF labelers' preferences and the reward model.

Putting some words in AGENTS.md hardly seems like the most annoying thing.

tip: Add a /fix command that tells it to fix $1 and then update AGENTS.md with the text that'd stop it from making that mistake in the future. Use your nearest LLM to tweak that prompt. It's a good timesaver.

Re:My AI Adoption Journey on Thursday February 05, 2026 @06:02PM (#46906658)
[Edit: I may have been replying to another comment in my head as now I re-read it and I'm not sure I've said the same thing as you have. Oh well.]

I agree. This is how I see it too. It's more like a shortcut to an end result that's very similar (or much better) than I would've reached through typing it myself.

The other day I did realise that I'm using my experience to steer it away from bad decisions a lot more than I noticed. It feels like it does all the real work, but I have to remember it's my/our (decades of) experience writing code playing a part also.

I'm genuinely confused when people come in at this point and say that it's impossible to do this and produce good output and end results.

Re:When internal hostnames are leaked to the clown on Thursday February 05, 2026 @10:20AM (#46900625)
How do you know?
Re:A Step Behind the Bleeding Edge: A Philosophy on AI in Dev on Saturday January 31, 2026 @01:26PM (#46839205)
You don't know what you're missing.
Re:The coming war on Car Ownership on Thursday January 29, 2026 @04:07PM (#46816592)
> All of US pay for these subsidies via significantly higher healthcare prices.

FTFY

Can you not see how you truly, deep down, are afraid you might be wrong?

It's clouding your vision.

The problem with replying to the proof-demanders is that they'll always pick it apart and find some reason it doesn't fit their definition. You must be familiar with that at this point.
Is this a joke? Are you genuinely implying that no one has ever got an LLM to write code that does exactly what they want?
Re:Web-based editor on Tuesday September 07, 2021 @08:39AM (#28443971)
> Considering nearly every S/W dev globally can make that in a couple of days

This is only possibly true if your definition of "global" is "1st world".

Re:Mighty Makes Google Chrome Faster on Thursday April 29, 2021 @07:45PM (#26988263)
> The cloud could be the worst thing that can happen to the Internet

Is this not the same as saying "The internet could be the worst thing that can happen to the Internet"?

Re:What's Cooking on Sourcehut? on Friday April 23, 2021 @05:06PM (#26919198)
I can't believe you're writing proprietary software. I thought you were principled.
I agree. I've learnt my lesson and I will _NOT_ be running "npm install" on any greenfield projects anymore. I like the idea of Svelte but I do not like its dependency on npm/node.

At least with Vue I can just include the library into the page, with no mandatory build step.

No problem. I hate the thought of data loss. They may be better now, but who knows, it's worth being sure.
Re:It’s time to stop using SMS for security on Tuesday March 16, 2021 @02:45PM (#26481091)
What was the conceptual change that occurred in 1995?
Re:Debian Running on Rust Coreutils on Thursday March 11, 2021 @08:19AM (#26423661)
I was sad to find out that rms apparently didn't even say it.
Re:Debian Running on Rust Coreutils on Thursday March 11, 2021 @08:16AM (#26423625)
I see. Interesting.
Re:Debian Running on Rust Coreutils on Thursday March 11, 2021 @04:10AM (#26421925)
What are you talking about? Where does he "suck up" all Justfiles?
Re:GitHub Source Code Leak on Friday November 06, 2020 @09:43AM (#25007891)
Why?
Re:Google users locked out after 15 years' use on Tuesday November 03, 2020 @06:06AM (#24978486)
> $110.15

> At minimum wage ($7.25[2]), a single domain name would take about half a month to earn

I don't understand.

110/7 is 15 hours, not 15 days.

Re:Google users locked out after 15 years' use on Tuesday November 03, 2020 @04:45AM (#24978051)
> If takeouts could be configured to download automatically through Google drive, that would be amazing.

They can. And Dropbox, and various other cloud providers.

Older »