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

AI adoption and Solow's productivity paradox

fortune.com

Post reply on HN
Print view View original

581–590 of 783 posts

Re: AI adoption and Solow's productivity paradox

#581

Post by simonw »

LunaSea wrote: post #503
simonw wrote: post #235

Earlier quoted context omitted.

Here are a few of mine from the past month - for all of them 90%+ of the code written by Claude Code: - https://github.com/simonw/sqlite-history-json - https://github.com/simonw/sqlite-ast - https://github.com/simonw/showboat - 292 stars - https://github.com/simonw/datasette-showboat - https://github.com/simonw/rodney - 290 stars and 4 contributors who aren't me or Claude - https://github.com/simonw/chartroom Noting…

I quickly read through the `sqlite-history-json` project and it's only a few hundred lines of code and the code doesn't use transactions which means that it can fail and leave the state of the code and database in an inconsistent state.

Being only a few hundred lines of code is a pro, not a con (it's 2,800 including tests: https://tools.simonwillison.net/sloccount?repo=https%3A%2F%2... - lines counted by my vibe-coded port of the classic Perl SLOCCount tool to run in a browser using Perl-in-WebAssembly.)

It does use transactions in the form of savepoints which means they can be nested: https://github.com/simonw/sqlite-history-json/blob/53e66b279...

Transactions are tested here: https://github.com/simonw/sqlite-history-json/blob/53e66b279...

I lead with sqlite-history-json because I think it's the most impressive of the bunch - it solves a difficult problem in an elegant way with code I would have been proud to write by hand.

Re: AI adoption and Solow's productivity paradox

#582

Post by docmars »

vidarh wrote: post #286
overgard wrote: post #170

Earlier quoted context omitted.

I don't think "results don't match promises" is the same as "not knowing how to use it". I've been using Claude and OpenAI's latest models for the past two weeks now (probably moving at about 1000 lines of code a day, which is what I can comfortably review), and it makes subtle hard-to-find mistakes all over the place. Or it just misunderstands well known design patterns, or does something bone headed. I'm fine with…

Do you use an agent harness to have it review code for you before you do? If not, you don't know how to use it efficiently . A large part of using AI efficiently is to significantly lower that review burden by having it do far more of the verification and cleanup itself before you even look at it.

This is correct, but part of the issue is that it significantly increases token usage costs. Some companies are doing:

- PRD and spec fulfillment review

- code review + correction loops

- security review + corrections

- addl. test coverage and tidying

- addl. type checks and tidying

- addl. lint checks and tidying

- maybe more I haven't listed

And these are run after each commit, so you can only imagine the costs per engineer doing this 10, 20, 50+ times per day depending on how much work they're knocking out.

Re: AI adoption and Solow's productivity paradox

#583

Post by LadyCailin »

simonw wrote: post #65
amrocha wrote: post #49

Earlier quoted context omitted.

There’s lots of slop out there, that doesn’t mean it’s actually good or useful code.

Keep moving those goal posts.

Even if this was goalpost moving, is it really an unreasonable ask to not have slop everywhere?

Re: AI adoption and Solow's productivity paradox

#585

Post by riskable »

vidarh wrote: post #286
overgard wrote: post #170

Earlier quoted context omitted.

I don't think "results don't match promises" is the same as "not knowing how to use it". I've been using Claude and OpenAI's latest models for the past two weeks now (probably moving at about 1000 lines of code a day, which is what I can comfortably review), and it makes subtle hard-to-find mistakes all over the place. Or it just misunderstands well known design patterns, or does something bone headed. I'm fine with…

Do you use an agent harness to have it review code for you before you do? If not, you don't know how to use it efficiently . A large part of using AI efficiently is to significantly lower that review burden by having it do far more of the verification and cleanup itself before you even look at it.

> Do you use an agent harness to have it review code for you before you do?

Right now you need to be Uncle Moneybags to do this in your personal life.

If you're lucky, your employer is footing the bill but otherwise... Ugh. It's like converting your app running perfectly fine on a cheap VPS to AWS Lambda. In theory, it's fine but in reality the next bill you get could make you faint.

Re: AI adoption and Solow's productivity paradox

#586

Post by Ithildin »

If I do something faster by pairing with AI, why should my employer reap the benefit? Why would I pass the savings on to my employer?

Could it be that employers are not seeing the difference because most employees are doing something else with the time they've saved by using AI?

There's been massive wage stagnation, benefits are crap, they play games with PTO. Most people I talk to who use AI as a part of their workflow are taking advantage of something nice that has come their way for a change.

Re: AI adoption and Solow's productivity paradox

#587

Post by docmars »

LunaSea wrote: post #418
XenophileJKO wrote: post #397

Earlier quoted context omitted.

If you know good architecture and you are testing as you go, I would say, it is probably pretty damn close to being able to build a company without looking at the code. Not without "risk" but definitely doable and plausible. My current project that I started this weekend is a rust client server game with the client compiled into web assembly. I do these projects without reading the code at all as a way to gauge what…

Security auditor and criminals have a bright future ahead of them.

True, but you'd be surprised how much you can tighten up a codebase by asking a heftier model to do a security review and suggest fixes.

Re: AI adoption and Solow's productivity paradox

#589

Post by riskable »

illiac786 wrote: post #569
whynotminot wrote: post #143

Earlier quoted context omitted.

It’s also pretty wild to me how people still don’t really even know how to use it. On hacker news, a very tech literate place, I see people thinking modern AI models can’t generate working code. The other day in real life I was talking to a friend of mine about ChatGPT. They didn’t know you needed to turn on “thinking” to get higher quality results. This is a technical person who has worked at Amazon. You can’t expec…

> On hacker news, a very tech literate place, I see people thinking modern AI models can’t generate working code. I am completely flooded with comments and stories about how great LLMs are at coding. I am curious to see how you get a different picture than this? Can you point me to a thread or a story that supports your view? At the moment, individuals thinking AI cannot generate working code seem almost inexistent t…

It's a real thing, but usually tied to IT folks that tried ChatGPT ~2 years ago (in a web browser) and had to "fix" whatever it output. That situation solidified their "understanding of AI" and they haven't updated their knowledge on the current situation (because... No pressing need).

Folks like this have never used AI inside of an IDE or one of the CLI AI tools. Without that perspective, AI seems mostly like a gimmick.

Re: AI adoption and Solow's productivity paradox

#590

Post by mghackerlady »

gzread wrote: post #368
aurareturn wrote: post #314

Earlier quoted context omitted.

I might not be a good programmer but I've been a very productive one. Someone who is good at writing code isn't always good at making money.

AI slop books made more money than JK Rowling, too.

To be fair, I wouldn't be entirely surprised if they were better than barfs onto a page. She's not exactly Tolkien
Post reply on HN