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

Let's talk about LLMs

b-list.org

Post reply on HN
Print view View original

181–190 of 201 posts

Re: Let's talk about LLMs

#181

Post by bonesss »

leptons wrote: post #84
kelnos wrote: post #27

>> Within just this group the ratios between best and worst performances averaged about 10:1 on productivity measurements and an amazing 5:1 on program speed and space measurements! > (although I’m personally skeptical of the “10x programmer” concept, the software industry overall does seem to accept it as true) To be fair, this statement from Brooks doesn't entirely match with the "10x programmer" we talk about. My…

There's no such thing as a "10x" programmer, and anyone who uses it doesn't know what they're talking about. 10x relative to what exactly? It's not a statement grounded in any kind of reality.

There are 10x developers, your assertions are overly broad and unanchored from the insights large developer stables can create.

10x relative to drumroll> other developers. Developers in the same place doing the same-ish things, only on metrics and outcomes you objectively have someone(s) making outputs that are outstripping teams and all their players.

Not 10x LoC, 10x problem to solution time, maintenance costs, and time/cost to create. At everything always? No, at relevant things. In fact, some of those nerds arguably go past that by being able to solve things the -4x to 5x’ers cannot. Any fulfillment is infinitely faster than non-fulfilment.

I’ve worked with several and have seen their projects numbers year on year in a large pool. SQL gurus who could get there inconceivably fast because their guruship let them conceive better. Independently created solutions that obviate existing systems and components, and got there >10x cheaper and were >10x cheaper to maintain.

Never outbid someone by planning to do way less, smarter and faster? 10x ain’t that much if the other dudes are average consultant houses.

Re: Let's talk about LLMs

#182

Post by aspenmartin »

nothinkjustai wrote: post #178
aspenmartin wrote: post #162

Earlier quoted context omitted.

Well I’m not offended but it sounds like you may not be paying attention? Do you know the capital outlay that has gone into infra buildouts? several people here have described “6 months” of AI mania—-the fact that people are saying 6 months is exactly the point. Development has been going on since 2010s. All of the “boosters” as HN likes to say have been saying “hey this thing is huge and the performance trends are s…

What does Siri have to do with anything? Honestly your posts read like satire. You’re treating LLMs like a religion and the release of Opus 4.6 as like some type of rapture. Idk man, if it’s some sort of bit or false flag thing well played, if not…well good luck.

What exactly do you find satirical?

- obviously LLMs are not a religion I’m using it to illustrate a point

- 5-6 months ago was when agent perf hit a meaningful inflection point where adoption has exploded. It’s why people in this thread reference “the past 6 months” whether or not they realize we’ve been on the same path for years now

So to overextend the metaphor, opus 4.5 was really kind of the right fit for the rapture.

I mean no need to take any of this seriously, I have worked on benchmarks and measurement in an AI lab professionally for over 4 years now, in software and data science for 8 and before got a PhD in Astro, like I’m not some sort of armchair person with no understanding of this field. Though I do find it entertaining when my background in an AI lab is people’s favorite reason to dismiss this :)

I find that when people find stuff like this satirical they often don’t really know the industry or underlying mechanics that well. Not saying that’s you but as ridiculous as I apparently sound to you, do consider that sounds even more ridiculous to not understand the tsunami that is coming right for you…

Re: Let's talk about LLMs

#183

Post by aspenmartin »

nothinkjustai wrote: post #177
handoflixue wrote: post #158

Earlier quoted context omitted.

If you want to move the goal posts, that's fine, acknowledge it: the original claim I'm responding to was "We're almost 6 months into all this AI-code madness" If you want to set GPT as the target, that's even easier! In that decade it has passed the Turing Test, solved novel open math problems, generates audio, video, and music, and can write coherent code. Again, there is no technology that has improved more rapidl…

I’d say the internet did, since it literally connected people across the globe in real time, which actually provided the technology that allows LLMs and other similar tech to exist in the first place. I think it’s pretty clear the internet has had 10x the impact of LLMs so far. Maybe 100x

The internet has been around since the 80s…ChatGPT came out 4 years ago. The internet took decades to build out the infrastructure. Inflation adjusted capex for AI infrastructure already far surpasses that of the internet. You’re talking about a technology that doesn’t just make things easier it replaces entire swaths of work. Under some weird measurement you may be right but I mean cmon.

Re: Let's talk about LLMs

#184

Post by aspenmartin »

leptons wrote: post #180
aspenmartin wrote: post #114

Earlier quoted context omitted.

Always amazes me how we’re on a platform with “ycombinator” in the url and people don’t understand how private companies scale to capture market share. You’re right Uber was that company that ran at a loss for so long and collapsed, another YOLO business strategy. Or maybe it was Amazon or…hmm I forget

I can't afford to take Ubers anymore. A trip that used to cost $7 now costs $40. AI is going to be the same to cover all the massive amounts of money already spent. You like your $200/mo plan now? How about when it's $2000/mo?

You are not at all wrong. Also get ready for the insidious advertising!

Re: Let's talk about LLMs

#185

Post by empthought »

soco wrote: post #147
empthought wrote: post #83

Earlier quoted context omitted.

You don't need a special feature for this. Just tell the coding assistant what to do.

Then watch it f'up half your codebase because it thinks it's slightly related to your examples. The alternative, giving it 10 examples, is actually more work.

I don’t think you’ve actually used any of these tools. 10 different examples in the same session would almost certainly make them perform worse.

Re: Let's talk about LLMs

#187

Post by lajsfoihasdf »

aspenmartin wrote: post #182
nothinkjustai wrote: post #178

Earlier quoted context omitted.

What does Siri have to do with anything? Honestly your posts read like satire. You’re treating LLMs like a religion and the release of Opus 4.6 as like some type of rapture. Idk man, if it’s some sort of bit or false flag thing well played, if not…well good luck.

What exactly do you find satirical? - obviously LLMs are not a religion I’m using it to illustrate a point - 5-6 months ago was when agent perf hit a meaningful inflection point where adoption has exploded. It’s why people in this thread reference “the past 6 months” whether or not they realize we’ve been on the same path for years now So to overextend the metaphor, opus 4.5 was really kind of the right fit for the r…

[dead]

Re: Let's talk about LLMs

#188

Post by wavemode »

handoflixue wrote: post #159
wavemode wrote: post #117

Earlier quoted context omitted.

You're scoring rhetorical points while talking past my entire comment. Hard to say if you even read it. I'm not "convincing" anyone of anything. I'm stating the reasons that I, personally, am unconvinced of a specific claim being made to me .

I mean, I quoted multiple passages and established why I think your logic is flawed. If you're convinced by bad logic, so be it.

If you read my entire comment and thought that showing me a benchmark chart remotely addresses the point I'm making, well... I don't know what to tell you.

Re: Let's talk about LLMs

#189

Post by marcus_holmes »

weakfish wrote: post #166
marcus_holmes wrote: post #94

I was waiting for the "so I tried coding something with an LLM myself, and I found..." paragraph. But apparently the author never did try it, or at least if they did, they didn't write about it. This is a very academic approach to the subject - read what other people have written about it without ever doing it yourself. Study what someone said about LLM coding 50 years ago, before they were even invented, to see what…

Are you asking, essentially, to move past the data and evidence and get anecdotes? That seems the opposite of useful, and tbh LLM coding has wayyyy too much anecdotal 'evidence' going on.

I'm not sure what "the evidence" is in this case?

I mean, we have lots of people using LLMs to write software in different ways, as we explore this space. I don't really see how "the evidence" can be different from "anecdotes" at this stage of the exploration?

There have been a couple of studies done on LLM-assisted dev vs non-LLM-assisted dev, but the author doesn't cite them.

Re: Let's talk about LLMs

#190

Post by marcus_holmes »

spopejoy wrote: post #179
marcus_holmes wrote: post #130

Earlier quoted context omitted.

I had to go re-read the article to make sure, but it doesn't address teams or scaling to teams at all, so I'm not sure why you're asking about that? The article is talking about inherent vs accidental complexity, amongst other points, and if the author had actually tried developing with an LLM, they might have worked out how LLM coding does address some of this.

- The DORA report is about organizations not individuals - Mythical man-month is about organizations not individuals - No Silver Bullet: "I believe the hard part of building software to be the specification, design, and testing of this conceptual construct, not the labor of representing it and testing the fidelity of the representation." Clearly he's NOT talking about the 10x dev building the whole thing themselves,…

Meh, I'll concede that Fred Brooks was mostly writing about developing software within an organisation, and therefore writing about teams.

Coding time is important if it gates experiments and spikes. If you have to work out your architecture on paper because actually coding it up is a serious expense, then it becomes harder to experiment with different designs. In an LLM world where coding time is very cheap, it becomes easier to experiment and try things out. Developing an entire architecture and then abandoning it because it turned out that it didn't scale too well, or couldn't handle some edge cases, is not a major mistake or problem any more. There's no pressure to keep old code because it cost a lot of money to develop. You can spike an entire system, decide that it was a useful experiment, but didn't work, delete the repo, and go get lunch. This is new, and important.

Post reply on HN