Live data from Hacker News

Everyone in Seattle hates AI

jonready.com

601–610 of 1001 posts

Re: Everyone in Seattle hates AI

#601

Earlier quoted context omitted.

Pretty much. Someone on our team put out a code review for some new feature and then bounced for a 2 week vacation. One of our junior engineers approved it. Despite the fact that it was in a section of dead code that wasn’t supposed to even be enabled yet, it managed to break our test environment. Took senior engineers a day to figure out how that was even possible before reverting. We had another couple engineers ta…

> One of our junior engineers approved it. pretty sure the process I've seen most places is more like: one junior approves, one senior approves, then the owner manually merges. so your process seems inadequate to me, agents or not. also, was it tagged as generated? that seems like an obvious safety feature. As a junior, I might be thinking: 'my senior colleague sure knows lots of this stuff', but all it would take to…

> pretty sure the process I've seen most places is more like: one junior approves, one senior approves, then the owner manually merges.

Yeah that’s what I think we need to enforce. To answer your question, it was not tagged as AI generated. Frankly, I think we should ban AI-generated code outright, though labeling it as such would be a good compromise.

Re: Everyone in Seattle hates AI

#602

Earlier quoted context omitted.

That is the problem. AI is optimized to solve a problem no matter what it takes. It will try to solve one problem by creating 10 more. I think long time/term agentic AI is just snake oil at this point. AI works best if you can segment your task into 5-10 minutes chunks, including the AI generating time, correcting time and engineer review time. To put it another way, a 10 minute sync with human is necessary, otherwis…

> it just makes software engineering into bothering supervisor job. I'm pretty sure this is the entire enthusiasm from C-level for AI in a nutshell. Until AI SWE resisted being mashed into a replaceable cog job that they don't have to think/care about. AI is the magic beans that are just tantalizingly out of reach and boy do they want it.

Luckily for us, technologies like SQL made similar promises (for more limited domains) and C suites couldn't be bothered to learn that stuff either.

Ultimately they are mostly just clueless, so we will either end up with legions of way shittier companies than we have today (because we let them get away with offloading a bunch of work to tools they rms int understand and accepting low quality output) or we will eventually realize the continued importance of human expertise.

Re: Everyone in Seattle hates AI

#603

Earlier quoted context omitted.

My friends at Google are some of the most negative about the potential of AI to improve software development. I was always surprised by this and assumed internally at Google would be one of the first places to adopt these.

Google has good engineers. Generally I've noticed the better someone is at coding the more critical they are of AI generated code. Which make sense honestly. It's easier to spot flaws the more expert you are. This doesn't mean they don't use AI gen code, just they are more careful with when an where.

IMO this is mostly just an ego thing. I often see staff+ engineers make up reasons why AI is bad, when really it’s just a prompting skill issue.

When something threatens a thing that gives you value, people tend to hate it

Re: Everyone in Seattle hates AI

#604

Earlier quoted context omitted.

The thing that changed my view on LLMs was solo traveling for 6 months after leaving Microsoft. There were a lot of points on the trip where I was in a lot of trouble (severe food sickness, stolen items, missed flights) where I don't know how I would have solved those problems without chatGPT helping.

This is one of the most depressing things I have ever read on Hacker News. You claim to have become so de-actualized as a human being that you cannot fulfill the most basic items of Maslow’s Hierarchy of Needs (food, health, personal security, shelter, transportation) without the aid of an LLM.

Relax

Re: Everyone in Seattle hates AI

#605

Earlier quoted context omitted.

Yeah, "Engineers don't try" is a frustrating statement. We've all tried generative AI, and there's not that much to it — you put text in, you get text back out. Some models are better at some tasks, some tools are better at finding the right text and connecting it to the right actions, some tools provide a better wrapper around the text-generation process. Certain jobs are very easy for AI to do, others are impossibl…

I’ve been an engineer for 20 years, for myself, small companies, and big tech, and now working for my own saas company. There are many valid critiques of AI, but “there’s not much there” isn’t one of them. To me, any software engineer who tries an LLM, shrugs and says “huh, that’s interesting” and then “gets back to work” is completely failing at their actual job, which is using technology to solve problems. Maybe AI…

I would've thought that in 20 years you would have met other devs who do not think like you?

something I enjoy about our line of work is there are different ways to be good at it, and different ways to be useful. I really enjoy the way different types of people make a team that knows its strengths and weaknesses.

anyway, I know a few great engineers who shrug at the agents. I think different types of thinker find engagement with these complex tools to be a very different experience. these tools suit some but not all and that's ok

Re: Everyone in Seattle hates AI

#606

Earlier quoted context omitted.

My friends at Google are some of the most negative about the potential of AI to improve software development. I was always surprised by this and assumed internally at Google would be one of the first places to adopt these.

I've generally found an inverse correlation between "understands AI" and "exuberance for AI". I'm the only person at my current company who has had experience at multiple AI companies (the rest have never worked on it in a production environment, one of our projects is literally something I got paid to deliver customers at another startup), has written professionally about the topic, and worked directly with some big…

I think there is a correlation between when you can you expect from something when I know their internals vs someone that doesn’t know but is not like who knows internals is much much better.

Example: many people created websites without a clue of how they really work. And got millions of people on it. Or had crazy ideas to do things with them.

At the same time there are devs that know how internals work but can’t get 1 user.

pc manufacturers never were able to even imagine what random people were able to do with their pc.

This to say that even if you know internals you can claim you know better, but doesn’t mean it’s absolute.

Sometimes knowing the fundamentals it’s a limitation. Will limit your imagination.

Re: Everyone in Seattle hates AI

#607

Earlier quoted context omitted.

My friends at Google are some of the most negative about the potential of AI to improve software development. I was always surprised by this and assumed internally at Google would be one of the first places to adopt these.

I've generally found an inverse correlation between "understands AI" and "exuberance for AI". I'm the only person at my current company who has had experience at multiple AI companies (the rest have never worked on it in a production environment, one of our projects is literally something I got paid to deliver customers at another startup), has written professionally about the topic, and worked directly with some big…

This completely explains why so many engineers are skeptical of AI while so many managers embrace it: The engineers are the ones who understand it.

(BTW, if you're an engineer who thinks you don't understand AI or are not qualified to work on it, think again. It's just linear algebra, and linear algebra is not that hard. Once you spend a day studying it, you'll think "Is that all there is to it?" The only difficult part of AI is learning PyTorch, since all the AI papers are written in terms of Python nowadays instead of -- you know -- math.)

I've been building neural net systems since the late 1980s. And yes they work and they do useful things when you have modern amounts of compute available, but they are not the second coming of $DEITY.

Re: Everyone in Seattle hates AI

#608

Ex-Google here; there are many people both current and past-Google that feel the same way as the composite coworker in the linked post. I haven't escaped this mindset myself. I'm convinced there are a small number of places where LLMs make truly effective tools (see: generation of "must be plausible, need not be accurate" data, e.g. concept art or crowd animations in movies), a large number of places where LLMs make…

Is it true that it's bad for learning new skills? My gut tells me it's useful as long as I don't use it to cheat the learning process and I mainly use it for things like follow up questions.

I think what it comes down to, and where many people get confused, is separating the technology itself from how we use it. The technology itself is incredible for learning new skills, but at the same time it incentivizes people to not learn. Just because you have an LLM doesn't mean you can skip the hard parts of doing textbook exercises and thinking hard about what you are learning. It's a bit similar to passively watching youtube videos. You'd think that having all these amazing university lectures available on youtube makes people learn much faster, but in reality in makes people lazy because they believe they can passively sit there, watch a video, do nothing else, and expect that to replace a classroom education. That's not how humans learn. But it's not because youtube videos or LLMs are bad learning tools, it's because people use them as mental shortcut where they shouldn't.

Re: Everyone in Seattle hates AI

#609

Earlier quoted context omitted.

Some people really do hate AI, it's not entirely about the layoffs. This is a well insulated bubble but you can find tons of anti-AI forums online.

Some people take find their life meaning through craft and work. When that craft is suddenly less scarce, less special, so does that craft-tied meaning. I wonder if these feelings are what scribes and amanuenses felt when the printing press arrived. I do enjoy programming, I like my job and take pride on it, but I actively try for it not to be the life-mean giving activity. I'm a just mercenary of my trade.

The craft isn't any less scarce. If anything, only more. The craft of building wooden furniture is just as scarce as ever, despite the existence of Ikea.

Re: Everyone in Seattle hates AI

#610

Earlier quoted context omitted.

Well yeah. And because when an expert looks at the code chatgpt produces, the flaws are more obvious. It programs with the skill of the median programmer on GitHub. For beginners and people who do cookie cutter work, this can be incredible because it writes the same or better code they could write, fast and for free. But for experts, the code it produces is consistently worse than what we can do. At best my pride dem…

> It programs with the skill of the median programmer on GitHub This is a common intuition but it's provably false. The fact that LLMs are trained on a corpus does not mean their output represents the median skill level of the corpus. Eighteen months ago GPT-4 was outperforming 85% of human participants in coding contests. And people who participate in coding contests are already well above the median skill level on…

I don't think this disproves my claim, for several reasons.

First, I don't know where those human participants came from, but if you pick people off the street or from a college campus, they aren't going to be the world's best programmers. On the other hand, github users are on average more skilled than the average CS student. Even students and beginners who use github usually don't have much code there. If the LLMs are weighted to treat every line of code about same, they'd pick up more lines of code from prolific developers (who are often more experienced) than they would from beginners.

Also in a coding contest, you're under time pressure. Even when your code works, its often ugly and thrown together. On github, the only code I check in is code that solves whatever problem I set out to solve. I suspect everyone writes better code on github than we do in programming competitions. I suspect if you gave the competitors functionally unlimited time to do the programming competition, many more would outperform GPT-4.

Programming contests also usually require that you write a fully self contained program which has been very well specified. The program usually doesn't need any error handling, or need to be maintained. (And if it does need error handling, the cases are all fully specified in the problem description). Relatively speaking, LLMs are pretty good at these kind of problems - where I want some throwaway code that'll work today and get deleted tomorrow.

But most software I write isn't like that. And LLMs struggle to write maintainable software in large projects. Most problems aren't so well specified. And for most code, you end up spending more effort maintaining the code over its lifetime than it takes to write in the first place. Chatgpt usually writes code that is a headache to maintain. It doesn't write or use local utility functions. It doesn't factor its code well. The code is often overly verbose. It often writes code that's very poorly optimized. Or the code contains quite obvious bugs for unexpected input - like overflow errors or boundary conditions. And the code it produces very rarely handles errors correctly. None of these problems really matter in programming competitions. But it does matter a lot more when writing real software. These problems make LLMs much less useful at work.

Post reply on HN