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

Don't be a meat proxy

gruhn.me

Post reply on HN
Print view View original

611–620 of 786 posts

Re: Don't be a meat proxy

#611

Post by dontparticipate »

We've already started firing people for being meat proxies at my company. Was reviewing a huge and awful slop code base that a vibe coder only had been working on getting started. I started giving my feedback to the developer, and it became quickly clear that I was the only human involved in developing this "software" and the only one that had ever read the code and evaluated it.

Cost the company a ton and had to take it over from scratch. AI is unleashing a small army of 0.01x engineers and making them harder to catch, so not only are you burning the tokens, you're also setting that labor cost on fire.

Re: Don't be a meat proxy

#612

Post by projectazorian »

eddythompson80 wrote: post #8

I deal with this all day long at work and it’s exhausting. People almost acting like no one has thought of it “I asked Claude what happened, and it spit out this 300 line response. Can you read it for me and see if it’s right?” What kills me is you might expect this from a busy high level manager that doesn’t really understand the technical details and they just point the AI to an error they got. They don’t know how…

I'm not opposed to straight copypasta, but it should be limited to excerpts only or at the very least workshopped with the LLM to reduce verbosity. Certainly it should at least be fully read and grokked.

I can count on one hand the number of times I've seen LLMs one-shot responses that are fit to paste to Slack verbatim. It does happen though.

Re: Don't be a meat proxy

#613

Post by hirsin »

ltbarcly3 wrote: post #516
rufasterisco wrote: post #494

Earlier quoted context omitted.

Any non trivial coding task requires a conversation between human and ai. Just have devs commit (or provide however you like) their ai coding sessions. It then becomes obvious if someone has, or has not, reviewed the code, because then they discussed it. Bonus points: on very complex tasks, you can more or less resume the conversation, via session files. Same for schoolwork, research papers, and a lot more. The conve…

I want to suggest you read https://www.amazon.com/Advantage-Disadvantage-History-Hacket... You can't just remember every conversation and ... something useful. Indiscriminate memory buries you. What you need is "monumental" memories, which is also known as documentation. AI can be very effective in helping you manage that. That should be kept tight, just like code. You want to keep enough of a record of how you got t…

While I generally agree for general human living, I've seen impressive "digital hoarder" success using AI because it changes the dimensions of reasonableness.

Imagine your hoarder scenario except that the hoarder has unlimited storage space and a team of librarian/warehouse workers to go find and fetch the items needed for the wall hanging task.

Re: Don't be a meat proxy

#614

Post by mikebenfield »

ungreased0675 wrote: post #413
prymitive wrote: post #65

Earlier quoted context omitted.

I think a lot of that is people who were always very insecure because they’re mediocre engineers. Previously asking questions or not understanding something was a bit painful but now you can hear “you’re absolutely right!” and make progress all day every day. Until you need to interact with actual humans and that’s why you try to minimise it, hiding behind ai generated content.

I’m not a developer, but when I watch devs whose work process is prompt, copy-paste, try to run, paste error into code, try to run, etc. I can’t help but think they’re unskilled. There’s no brain engagement, no understanding of the bigger picture, just being a worse slower agent.

I think I’m pretty skilled, or at least I was, but much of my work now looks like this. The fact is in many cases Claude can diagnose and fix the issue quicker than I can even read it. It would be crazy not to take advantage of this. Of course it does get hard to resist the temptation to just become ever lazier over time.

Re: Don't be a meat proxy

#615

Post by skydhash »

embedding-shape wrote: post #555
skydhash wrote: post #522

Earlier quoted context omitted.

> Or just that you've never seen it, but you're open to it existing? This one. Only a sith deals in absolute. I don’t mind experiments to try to find methodologies for those tools. And I believe there are instances where they’ve been successfully used. The issue I have is the kind of generic statements that they are good enough to replace currently established methodologies. Like using AI is a panacea. > Personally,…

> Have you investigated how it has helped you? And if there are other, more economical way to get the same result? No I haven't, but I'm happy to just freeform walk you through my thinking on it: I typically write (wrote?) software for two purposes: consulting/freelancing for others so building what others want, or for simplifying and making my own life easier and more enjoyable. "Stupid" stuff like Home Assistant fo…

> What are the exact "currently established methodologies" you're talking about that cannot be replaced by LLMs + a harness today, just as some examples? You're probably right that those exists, but I'm curious to hear what you think would be the most difficult to replace today.

I was explaining [0] under another post that programming is mostly translation works. You take a specs and you formalize it using code, like going from sketch to a proper engineering drawing. Software design is more creative, where you take a problem and then comes up with a solution (creating the specs). Software Engineering is ensuring that those two are done well enough while consuming the least resources.

So a program is always a formal system. It's also static. It will be executed by a computer which will actually have a tangible effect in the real world. That effect is what's valuable. The program is the seed which let us control that effect. Aka it's the map that let us plan the journey, but it's not the territory that we will have to travel in.

The issue I keep pointing in most of my comment is thinking that the map is the territory. That the novel are the words and not the story so we need more words. Or that the code is more important than the user' workflows, se we are adding more buggy code, while not ensuring that the workflows are undisturbed.

> Nowadays, my entire home-lab is configured with Nix and almost everything except my workstation and some random stuff, runs NixOS. Everything is hosted on a local Forgejo instance, which also has it's own (custom "written" of course) agent acting on issues and PRs, and I have my harness basically maintain my entire home lab at this point.

It's also highlighted here where you focus more on the process than the output here. The goal is to have a working homelab. NixOS managing it is only the process (accidental complexity). If it's where truly about the goal and not NixOS and using AI, by this point, adding new nodes (software, devices,...) should be as easy as selecting it and adding it to the current system, like a strategy game.

You can see that philosophy in OpenBSD, where the focus is to have a working OS, not to work on developing an OS. A lot of software are done and it's mostly just bug fixing every once in a while. You can also see the same attitude in industrial engineering where you develop a solution and then use it for years. You don't spend all your time tweaking it and thus disturbing the production flow.

So yes, when I see a LLM methodology, it's mostly about the work itself, not the output of the work. There is no definition of done or even the idea of having one. It's work for the purpose of working.

[0] https://news.ycombinator.com/item?id=49157720

Re: Don't be a meat proxy

#616

Post by wuliwong »

dontparticipate wrote: post #611

We've already started firing people for being meat proxies at my company. Was reviewing a huge and awful slop code base that a vibe coder only had been working on getting started. I started giving my feedback to the developer, and it became quickly clear that I was the only human involved in developing this "software" and the only one that had ever read the code and evaluated it. Cost the company a ton and had to tak…

I was a big distributed, event-driven service guy for years. We used to talk about how it can make things 'faster' but you have to put some infrastructure around it in order to be able to move like that. I'm basically talking about CI/CD stuff, automated tests, security, load test, etc. You use feature flags, blue green deploys, stuff like that.

I _think_ that the same concept could apply to agentic code. If you have the guardrails in place, you could trust it a bit more. Anyway, that's a lot of what I think about and but have had limited success getting people to buy in as much as I think they should. :)

Re: Don't be a meat proxy

#618

Post by throw310822 »

AnimalMuppet wrote: post #579
throw310822 wrote: post #569

Earlier quoted context omitted.

> Not every question can be answered accurately by Claude You are assuming that people that send you a Claude answer would send you any Claude answer. This is your assumption- basically a negative prejudice towards those who use LLMs- and has nothing to do with sending a Claude answer per se. Do you understand the difference?

I do. I'm pretty sure that at least some humans will shove any Claude answer at you. And some won't. You are assuming that I am making an assumption. Your assumption is wrong. I am recognizing that some humans are lazy and will do the minimum, and the minimum is to type in some query and copy-paste the results. If you like, yes, I am prejudiced - prejudiced against that behavior. (I mean, I still trust a human who kn…

But then we agree. Some humans will behave that way. Then direct your anger against those who do so, instead of against those (the majority) who ask the questions, read and evaluate the answers, and forward them to you only if they think they might be correct and helpful.

Re: Don't be a meat proxy

#619

Post by pixl97 »

tomtomtom777 wrote: post #542
ungreased0675 wrote: post #413

Earlier quoted context omitted.

I’m not a developer, but when I watch devs whose work process is prompt, copy-paste, try to run, paste error into code, try to run, etc. I can’t help but think they’re unskilled. There’s no brain engagement, no understanding of the bigger picture, just being a worse slower agent.

I think the worst is when they are skilled. You're basically describing my work day. My job went from interesting, highly skilled work to mostly horrible brain rot over the past year. It's depressing and I don't know a way out. Other then a complete career change, throwing away 30 years of experience and a matching salary.

I mean, we're just the neo-luddites in this case. Or maybe the rust belt when jobs moved elsewhere.

The real question isn't if things are going to change around our jobs, its are we going to be able to move fast enough to avoid starving in the streets?

Re: Don't be a meat proxy

#620

Post by ImPostingOnHN »

throw310822 wrote: post #207
jagenabler2 wrote: post #173

Earlier quoted context omitted.

> hey, I don’t know the answer to your question but I went through Claude and got this response. Let me know if that answers your question otherwise we can dig deeper together That’s all you have to add to make this into a much more pleasant interaction

I agree. However, since I work in a small team and everyone knows me, I don't need to provide this entire context every time. I usually just prepend "according to Claude ..." and try to provide a one line recap of the main point. Then paste the relevant section of Claude's answer (two to three paragraphs).

Your response misses the words most critical to this interaction: "I don't know"
Post reply on HN