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

Software is mostly all you need

softwarefordays.com

Post reply on HN
Print view View original

31–40 of 52 posts

Re: Software is mostly all you need

#31

Post by jbmilgrom »

daxfohl wrote: post #25
jbmilgrom wrote: post #16

Earlier quoted context omitted.

Author here We are building this at docflowlabs ie a self-healing system that can respond to customer feedback automatically. And youre right that not all customers know what they want or even how to express it when they do, which is why the agent loop we have facing them is way more discovery-focused than the internal one. And we currently still have humans in the loop for everything (for now!) - e.g, the agent does…

Cool, I tried something similar over a couple weeks but the problem I ran into was that beyond a fairly low level of complexity, the English spec became more confusing than the code itself. Even for a simple multi-step KYC workflow, it got very convoluted and hard to make it precise, whereas in code it's a couple loops and if/else blocks with no possibility of misinterpretation. Have you encountered that at all, or h…

that's right, and agents turning specs into software can go in all sorts of directions especially when we don't control the input.

what we've done to mitigate is essentially backing every entrypoint (customer comment, internal ticket, etc) with a remote claude code session with persistent memory - that session essentially becomes the expert in the case. And we've developed checkpoints that work from experience (e.g. the root cause one) where a human has the opportunity to take over the wheel so to speak and drive in a different direction with all the context/history up to that point.

basically, we are creating a assembly line where agents do most of the work and humans increasingly less and less as we continue to optimize the different parts of assembly

as far as techniques, it's all boring engineering

* Temporal workflow for managing the lifecycle of a session

* complete ownership of the data model e2e. we dont use Linear for example; we built our own ticketing system so we could represent Temporal signals, github webhooks and events from the remote claude sessions exactly how we wanted

* incremental automation gains over and over again. We do a lot of the work manually first (like old fashioned hand coding lol) before trying to automate so we become experts in that piece of the assembly line and it becomes obvious how to incrementally automate...rinse and repeat

Re: Software is mostly all you need

#32

Post by jbmilgrom »

fooker wrote: post #30
jbmilgrom wrote: post #7

Earlier quoted context omitted.

> Perhaps the key to transparent/interpretable ML is to just replace the ML model with AI-coded traditional software and decision trees. This way it's still fully autonomously trained but you can easily look at the code to see what is going on. For certain problems I think thats completely right. We still are not going to want that of course for classic ML domains like vision and now coding, etc. But for those domain…

> We still are not going to want that of course for classic ML domains like vision It could make sense to decompose one large opaque model into code with decision trees calling out to smaller models having very specific purposes. This is more or less science fiction right now, 'mixture of experts' notwithstanding. You could potentially get a Turing award by making this work for real ;)

woah that would be crazy

Re: Software is mostly all you need

#34

Post by esafak »

raincole wrote: post #23
PostOnce wrote: post #22

Earlier quoted context omitted.

Tell that to all the OCR fuckups I see in all the ebooks I read.

Your ebooks are made with handwriting recognition...? What do you read, the digital version of Dead Sea Scrolls?

They could be OCRs of scanned printed books.

Re: Software is mostly all you need

#35

Post by daxfohl »

jbmilgrom wrote: post #31
daxfohl wrote: post #25

Earlier quoted context omitted.

Cool, I tried something similar over a couple weeks but the problem I ran into was that beyond a fairly low level of complexity, the English spec became more confusing than the code itself. Even for a simple multi-step KYC workflow, it got very convoluted and hard to make it precise, whereas in code it's a couple loops and if/else blocks with no possibility of misinterpretation. Have you encountered that at all, or h…

that's right, and agents turning specs into software can go in all sorts of directions especially when we don't control the input. what we've done to mitigate is essentially backing every entrypoint (customer comment, internal ticket, etc) with a remote claude code session with persistent memory - that session essentially becomes the expert in the case. And we've developed checkpoints that work from experience (e.g.…

Ooh, it sounds like you've already got most of the groundwork done for something I was wondering about yesterday: I'd love it if there was some way during an incident, for some system to pull all the PRs included in the latest release, check which agents worked on them (i.e. line in the commit message with an identifier that corresponds to the agent's LLM context and any other data at the time of commit), "rehydrate" these agents from the corresponding stored context, feed them the relevant incident data, and ask if it could be related to their changes and what to do about it.

In most cases it might not be much more valuable than just looking through the diffs from scratch with a new agent, but there are probably going to be some cases where a rehydrated agent is like "Doh, I meant to do X but it looks like I hallucinated Y instead. Here's a PR to fix it!"

I know that's just a small piece of what you're doing, but I think it's something that would be valuable on its own, and soon something that is likely to be "standard infrastructure" for any company that does even a little agentic coding (assuming it works). It'd probably even be "required infrastructure" in regulated industries; the fact that all these agent contexts are ephemeral has to be a red flag from a regulatory perspective.

Re: Software is mostly all you need

#36

Post by isodev »

boogrpants wrote: post #17
isodev wrote: post #5

> Neural networks excel at judgment I don’t think they do. I think they excel at outputting echoes of their training data that best fit (rhyme with, contextually) the prompt they were given. If you try using Claude with an obscure language or use case, you will notice that effect even more - it will keep pulling towards things it knows that aren’t at all what’s asked or “the best judgement” for what’s needed.

> I think they excel at outputting echoes of their training data that best fit (rhyme with, contextually) the prompt they were given. Just like people who get degrees in economics or engineering and engage in such role-play for decades. They're often pretty bad at anything they are not trained on. Coincidentally, if you put a single American English speaker on a team of native German language speakers you will notice…

> Just like people

I understand that having model related vocabulary borrow similar words we use to describe human brains and cognition gets confusing. We are not the same, we don’t “learn” the same we certainly don’t use the knowledge we posses in the same way.

The major difference between an LLM and a human is that as a human, I can look at your examples (which sound solid at first glance) and choose to truly “reason” about them in a way that allows me to judge if they’re correct or even applicable.

Re: Software is mostly all you need

#37

Post by perfmode »

isodev wrote: post #36
boogrpants wrote: post #17

Earlier quoted context omitted.

> I think they excel at outputting echoes of their training data that best fit (rhyme with, contextually) the prompt they were given. Just like people who get degrees in economics or engineering and engage in such role-play for decades. They're often pretty bad at anything they are not trained on. Coincidentally, if you put a single American English speaker on a team of native German language speakers you will notice…

> Just like people I understand that having model related vocabulary borrow similar words we use to describe human brains and cognition gets confusing. We are not the same, we don’t “learn” the same we certainly don’t use the knowledge we posses in the same way. The major difference between an LLM and a human is that as a human, I can look at your examples (which sound solid at first glance) and choose to truly “reas…

how’s your reasoning different from LLM reasoning?

Re: Software is mostly all you need

#38

Post by boogrpants »

isodev wrote: post #36
boogrpants wrote: post #17

Earlier quoted context omitted.

> I think they excel at outputting echoes of their training data that best fit (rhyme with, contextually) the prompt they were given. Just like people who get degrees in economics or engineering and engage in such role-play for decades. They're often pretty bad at anything they are not trained on. Coincidentally, if you put a single American English speaker on a team of native German language speakers you will notice…

> Just like people I understand that having model related vocabulary borrow similar words we use to describe human brains and cognition gets confusing. We are not the same, we don’t “learn” the same we certainly don’t use the knowledge we posses in the same way. The major difference between an LLM and a human is that as a human, I can look at your examples (which sound solid at first glance) and choose to truly “reas…

Obviously. You are not exactly the same as your nearest neighbor but have similar observable traits to outside observers.

But since you end up trying to differentiate yourself from an LLM in vague, conceptual qualifiers, not empirical differences, what it means to "reason" ...I am left uncertain what you mean at all.

An LLM can reject false assertions and generate false positives just like a human.

Within a culture too individual people become pretty copy paste distillations of their generations customs. As a social creature you aren't that different. Really all that sets you apart from other people or a computer is a unique meat suit.

Unfortunately for your meat suit most people don't care it exists and will carry on with their lives never noticing it.

While LLMs have massive valuations right now. Pretty sure the public has spoken when it comes to the differences you fail to illustrate actually mattering.

Re: Software is mostly all you need

#39

Post by obirunda »

perfmode wrote: post #37
isodev wrote: post #36

Earlier quoted context omitted.

> Just like people I understand that having model related vocabulary borrow similar words we use to describe human brains and cognition gets confusing. We are not the same, we don’t “learn” the same we certainly don’t use the knowledge we posses in the same way. The major difference between an LLM and a human is that as a human, I can look at your examples (which sound solid at first glance) and choose to truly “reas…

how’s your reasoning different from LLM reasoning?

What humans are known to do, and apparently there is no limit to what they won't, is anthropomorphizing. I think there's not been a single one of these discussions where someone inevitably says LLM's don't do X as well as a human and someone interjects in cult-like fashion.

Re: Software is mostly all you need

#40

Post by isodev »

obirunda wrote: post #39
perfmode wrote: post #37

Earlier quoted context omitted.

how’s your reasoning different from LLM reasoning?

What humans are known to do, and apparently there is no limit to what they won't, is anthropomorphizing. I think there's not been a single one of these discussions where someone inevitably says LLM's don't do X as well as a human and someone interjects in cult-like fashion.

slopbros need to keep vibing
Post reply on HN