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

Reliable Software in the LLM Era

quint-lang.org

Post reply on HN
Print view View original

31–39 of 39 posts

Re: Reliable Software in the LLM Era

#31

Post by forgetfreeman »

OutOfHere wrote: post #22
forgetfreeman wrote: post #17

Earlier quoted context omitted.

Not everything is worthy of a 5000 word Atlantic-style deconstruction when presented. I think the community largely embraces Hitchens' Razor. That being the case what is the minimum word count required to issue a dismissal?

Please. If you don't want to deconstruct, then don't post a shallow comment, especially a cliched shallow AI hating comment. As for Hitchen's razor, when it does apply, it applies implicitly, with no need for a comment or explicit mention.

I think not. Your position would have us choose between leaving obvious bullshit totally unchallenged or trying to sail upwind into the Bullshit Asymmetry Principle. I, for one, reject both options and am perfectly comfortable with others expressing their disdain succinctly. Tl;dr: pffft.

Re: Reliable Software in the LLM Era

#33

Post by OutOfHere »

forgetfreeman wrote: post #31
OutOfHere wrote: post #22

Earlier quoted context omitted.

Please. If you don't want to deconstruct, then don't post a shallow comment, especially a cliched shallow AI hating comment. As for Hitchen's razor, when it does apply, it applies implicitly, with no need for a comment or explicit mention.

I think not. Your position would have us choose between leaving obvious bullshit totally unchallenged or trying to sail upwind into the Bullshit Asymmetry Principle. I, for one, reject both options and am perfectly comfortable with others expressing their disdain succinctly. Tl;dr: pffft.

You will just have to hear the rule from the site admin, although the cost then to you could be higher.

It is not "bullshit" as you said it, since no real argument was made that it is. Hitchen's razor applies to your claim.

Re: Reliable Software in the LLM Era

#34

Post by 2001zhaozhao »

_pdp_ wrote: post #10

Nothing changes in terms of how to make reliable software. You need the same things like unit tests, integration tests, monitoring tools, etc. Basically AI now makes every product operate as if it has a vibrant open-source community with hundreds of contributions per day and a small core team with limited capacity.

> AI now makes every product operate as if it has a vibrant open-source community with hundreds of contributions per day and a small core team with limited capacity.

Hmm this is an interesting analogy. It suggests that we could design autonomous AI coding scaffold tools based on the patterns found in open source development communities.

Re: Reliable Software in the LLM Era

#35

Post by ptak_dev »

bugarela wrote: post #30
sastraxi wrote: post #9

The idea is interesting, but have some more respect for your potential readers and actually write the post. There’s so much AI sales drivel here it’s hard to see what’s interesting about your product. I’m more interested in the choices behind your design decisions than being told “trust me, it’ll work”.

Hey! We were not really sure how to pass on the information back when I wrote this in November, but since then we've packaged an opensourced all agents and AI stuff involved in that post: https://github.com/informalsystems/quint-llm-kit It's true what they say that it is easy to make a demo in AI, but super hard to turn demo into some product or thing other people can use. We are trying :) but also, most posts I read…

[flagged]

Re: Reliable Software in the LLM Era

#36

Post by sastraxi »

bugarela wrote: post #30
sastraxi wrote: post #9

The idea is interesting, but have some more respect for your potential readers and actually write the post. There’s so much AI sales drivel here it’s hard to see what’s interesting about your product. I’m more interested in the choices behind your design decisions than being told “trust me, it’ll work”.

Hey! We were not really sure how to pass on the information back when I wrote this in November, but since then we've packaged an opensourced all agents and AI stuff involved in that post: https://github.com/informalsystems/quint-llm-kit It's true what they say that it is easy to make a demo in AI, but super hard to turn demo into some product or thing other people can use. We are trying :) but also, most posts I read…

Yeah, I understand it's hard to find the balance here. I'd imagine you feel the need to ship at the same time as writing about it. For me, the post you shared in your reply as well as in the OP have been expanded to be about 3-4x as long as they need to be, and I'm going to assume you're using AI to generate them due to the writing style. My feedback is to consider writing shorter posts, but doing it by hand -- your prose style here is friendly, engaging and direct. I wish that your articles were the same.

TL;DR: I feel like my time is being wasted when I read AI-written articles, so I stop reading. Do with my anecdote what you will!

Re: Reliable Software in the LLM Era

#37

Post by shanjai_raj7 »

the part that is hard is when the model gets updated and your prompts behave differently. we dont always catch it in tests because the output still looks correct, just slightly off. by the time you notice something is wrong it has already been like that for a while.

Re: Reliable Software in the LLM Era

#38

Post by forgetfreeman »

OutOfHere wrote: post #33
forgetfreeman wrote: post #31

Earlier quoted context omitted.

I think not. Your position would have us choose between leaving obvious bullshit totally unchallenged or trying to sail upwind into the Bullshit Asymmetry Principle. I, for one, reject both options and am perfectly comfortable with others expressing their disdain succinctly. Tl;dr: pffft.

You will just have to hear the rule from the site admin, although the cost then to you could be higher. It is not "bullshit" as you said it, since no real argument was made that it is. Hitchen's razor applies to your claim.

"You will just have to hear the rule from the site admin, although the cost then to you could be higher."

I doubt that. My engagement with HN is on a level of morbid curiosity bordering on naked disdain. If n-gate was still a thing I probably would have quit checking the site altogether some years ago.

Re: Reliable Software in the LLM Era

#39

Post by bugarela »

sastraxi wrote: post #36
bugarela wrote: post #30

Earlier quoted context omitted.

Hey! We were not really sure how to pass on the information back when I wrote this in November, but since then we've packaged an opensourced all agents and AI stuff involved in that post: https://github.com/informalsystems/quint-llm-kit It's true what they say that it is easy to make a demo in AI, but super hard to turn demo into some product or thing other people can use. We are trying :) but also, most posts I read…

Yeah, I understand it's hard to find the balance here. I'd imagine you feel the need to ship at the same time as writing about it. For me, the post you shared in your reply as well as in the OP have been expanded to be about 3-4x as long as they need to be, and I'm going to assume you're using AI to generate them due to the writing style. My feedback is to consider writing shorter posts, but doing it by hand -- your…

Interesting, thank you! I feel the same about avoiding reading AI stuff, that's my worry here. In OP, I had way too much content and had AI help me reduce the size of if. But I learned a lot since then and some parts of the post bother me a bit now.

For the second post, it was a wall of prose that my colleague, Josef, wrote (I'm pretty sure without AI). I added the more concrete example sections and then asked AI to suggest some heading names to break down the prose, and I'm assuming this was my mistake as the headings are the first thing people read and AI loves headings. I always broke down text in headings before AI, this suck haha.

Anyway, I'll definitely increase my neurotic levels about this. It's not worth at all to rush things if it will make the post more empty to read in any way.

Post reply on HN