HNHacker News
TopNewBestAskShowJobs

andrei_says_

7,140 karma · joined December 18, 2015

submissionscomments
andrei_says_··on Ask HN: When and why did you start believing in God?
When I heard Carl Sagan defining “God” as the sum total of all matter, forces, causality chains, and events in the known and unknown universe. (Paraphrasing).

And when I read about the concept of all matter and all consciousness inhabiting each other just as they did prior and during the big bang.

andrei_says_··on Throwing AI-generated walls of text into conversations
“ I don't know, but here's my attempt to help.”

“I don’t know” would be so much more productive and respectful than “here, this may be useful or completely useless, figure it out on your own time. Took me 30 seconds to produce…”

A bit asymmetrical by time, signal to noise, and frankly, bs metrics.

andrei_says_··on GitHub confirms breach of 3,800 repos via malicious VSCode extension
You could, with a large enough captive audience.
andrei_says_··on Elon Musk has lost his lawsuit against Sam Altman and OpenAI
I sometimes wonder, what does one need a second 500 billion that the first 500 billion is not enough for?
andrei_says_··on Ask HN: What will happen as AI costs increase?
The lesson is that if you’re too big to fail no laws apply to you and there unlimited money to be made.

It has been learned very well.

The brazen violation of intellectual property was a precondition of making this technology useful. Taking the risk of breaking the law at this unprecedented scale was an informed decision made based on this very lesson.

andrei_says_··on Ask HN: Is there a term for feeling sad about forced AI adoption?
Also the replacement of craftsmanship with mass produced lower quality output possible for workers with less training and partial understanding to produce.

Processes are more efficient, machines are faster, workers are easily replaceable. The quality and complexity of the product is limited by these requirements.

andrei_says_··on Open Design: Use Your Coding Agent as a Design Engine
As someone who also designs, the few web page designs I saw produced by LLMs were ugly, generic, not accessible… unusable.

Even after prompting, I had to throw away all. No useful ideas just slop.

I’m sure there’s a lot to philosophize about a distant future when these work, but right now a waste of time.

andrei_says_··on Ask HN: Do you read differently now that anything could be AI generated?
People continue making amazing art and surely keep writing.

Slop is however flooding the space. I think at this point curation and trusting existing sources is crucial.

andrei_says_··on Ask HN: Which is Better–Android or iOS?
Better for privacy? iOS.

Better for voice search/assistant? Probably Android.

Better for voice dictation? Android.

Better keyboard? Android.

Better integration with other same-os devices? iOS.

Better UX? Likely whichever you’re used to.

Better for what?

andrei_says_··on Tell HN: I'm sick of AI everything
The equivalent of a student copying their homework.

Schools are not publishing houses for homework assignments. Struggling with the homework is the process training the student.

Was

andrei_says_··on Show HN: Mediator.ai – Using Nash bargaining and LLMs to systematize fairness
Do you use principles of nonviolent communication in your work? Or another framework to establish nondefensive listening?
andrei_says_··on Why don't we just ask AI to write assembler?
> If we generate so much code using AI that no one is really looking or reading the code anymore, just verifying end functionality, we can really just skip all that and go straight to assembler, no?

We could also just autogenerate the content of our websites, emails, contracts.

And we do, resulting in mountains of slop, varying from soulless to wildly incorrect.

Code is a precise way to describe intent. Using LLMs make up some of the intent results in the author not knowing what the precise functionality of the resulting code is.

The companies selling LLM services present this as magic which will magically do what the author wants it to do, without even the author themselves knowing or defining it.

In reality it is simply ignorance and lies.

Sorry we can’t wishful think good working software into existence.

andrei_says_··on Ask HN: Are Web Agencies Cooked?
Web agencies talk to clients and provide them with packaged services in understandable language.

Coding is a part of what clients pay for.

Agency clients want nothing to do with coding.

So make that part cheaper and faster and use the competitive advantage.

andrei_says_··on JSON formatter Chrome plugin now closed and injecting adware
I wonder if there’s a central repository of now exploited chrome extensions?

The chrome team does not seem to see security as a high enough priority.

andrei_says_··on Ask HN: Any interesting niche hobbies?
What are the meetup themes? What brings the interesting people in?
andrei_says_··on Write Your Own Copy
I’m genuinely impressed by the speed of mass slopification raising the toxicity of the whole internet.

And now poisoning human attention and the very ability to reason. At scale.

Makes me think of Kali Yuga, the age of ignorance.

Are we doing this to ourselves? Is it being done to us? Is it some kind of self destruct implanted by aliens in a dark forest scenario?

It’s just… really fascinating.

Also reminds me of this - story

https://www.lightspeedmagazine.com/fiction/different-kinds-o...

andrei_says_··on Anatomy of the .claude/ folder
Bikeshedding seems to have shifted from code to LLMs which is a step further.

We used to have the very difficult task of producing working scalable maintainable code describing complex systems which do what we need them to do.

Now on top of it we have the difficult task of producing this code using constantly mutating complex nondeterministic systems.

We are the circus bear riding a bicycle on a high wire now being asked to also spin plates and juggle chainsaws.

Maybe singularity means that time sunk into managing LLMs is equal to time needed to manually code similar output in assembly or punch cards.

andrei_says_··on Vibecoders Can't Build for Longevity
My dentist is more skilled at dentistry than 99.999% of humans. Why would I want to replace her with a machine which is more skilled than 99% of non dentists but 1000 times worse by an educated experienced dentist?
andrei_says_··on Vibecoders Can't Build for Longevity
Stupid but industrious and prolific. Also very confident.
andrei_says_··on Reports of code's death are greatly exaggerated
The question is, who made the new framework? Was it vibe coded by someone who does not understand its code?
andrei_says_··on Ask HN: What do you look for in your first 10 hires?
Care, initiative, long-term thinking, clear communication, collaboration skills, soft skills, candor, technical expertise.
andrei_says_··on What 81,000 people want from AI
An infinitely patient and somewhat schizophrenic teacher.
andrei_says_··on Ask HN: How is AI-assisted coding going for you professionally?
For the time being. Their manager is under constant pressure to lay them off and replace them with “ai”.
andrei_says_··on Ask HN: How is AI-assisted coding going for you professionally?
Which is fine if we are fine with pretending to do our jobs.

But deep down I know that slop is noise and words no longer represent understanding.

andrei_says_··on AirPods Max 2
That’s 385g vs Bose QC45’s 240g (8.5oz).
andrei_says_··on Tell HN: AI tools are making me lose interest in CS fundamentals
Important to remember that these skills evolve iteratively and often best in an apprenticeship environment. No one started with these abilities. They get developed by solving problems and getting better at it.
andrei_says_··on Shall I implement it? No
What ai? LLMs are language models, operating on words, with zero understanding. Or is there a new development which should make me consider anthropomorphizing them?
andrei_says_··on Show HN: DD Photos – open-source photo album site generator (Go and SvelteKit)
Turns out for a photo gallery json is just a bit tiny-er than html. And static html is so much simpler.

With proper links just working with the browsers built-in nav.

andrei_says_··on Don't trust AI agents
> Leaders haven't all gone insane, they answer to difficult questions with the narrative of least resistance.

Brilliantly said. I’d like to add - a distorted narrative actively, intentionally established and maintained by the entities profiting from the technology. Quite similar to the crypto scam hype cycle.

andrei_says_··on Don't trust AI agents
Maybe it depends on whose office? C-suite management who salivate after reducing software engineer headcount?
← PreviousPage 2 of 34Next →