The problem is that code as a 1-off is excellent, but as a maintainable piece of code that needs to be in source control, shared across teams, follow standard SLDC, be immutable, and track changes in some state - it's just not there.
If an intern handed me code like this to deploy an EC2 instance in production, I would need to have a long discussion about their decisions.
No, have you? They always seem to be missing from these types of posts. Personally I am skeptical, as AI has been abysmal at 1 shot provisioning actual quality cloud infrastructure. I wish it could, because it would make my life a lot less annoying. Unfortunately I have yet to really see it.
No, they're not. People talk about LLM-generated code the same way they talk about any code they're responsible for producing; it's not in fact the norm for any discussion about code here to include links to the code.
But if you're looking for success stories with code, they're easy to find.
> it's not in fact the norm for any discussion about code here to include links to the code.
I certainly didn't interpret "these types of posts" to mean "any discussion about code", and I highly doubt anyone else did.
The top-level comment is making a significant claim, not a casual remark about code they produced. We should expect it to be presented with substantiating artifacts.
I guess. I kind of side-eyed the original one-shotting claim, not because I don't believe it, but because I don't believe it matters. Serious LLM-driven code generation runs in an iterative process. I'm not sure why first-output quality matters that much; I care about the outcome, not the intermediate steps.
So if we're looking for stories about LLMs one-shotting high-quality code, accompanied by the generated code, I'm less sure of where those examples would be!
I could write a blog post exactly like this with my chatGPT history handy. That wasn't the point I was making. I am extremely skeptical of any claims that say someone can 1 shot quality cloud infrastructure without seeing what they produced. I'd even take away the 1-shot requirement - unless the person behind the prompt knows what they're doing, pretty much every example I've seen has been terrible.
I mean, I agree with you that the person behind the prompt needs to know what they're doing! And I don't care about 1-shotting, as I said in a sibling comment, so if that's all this is about, I yield my time. :)
There are just other comments on this thread that take as axiomatic that LLM-generated code is bad. That's obviously not true as a rule.
Like many things it’s a combination of many answers, but from my own experience it’s a simpler pipeline with Apple phones. On Android it’s supporting tons of screen sizes, out of date versions, different vendor modifications to the OS.
Looking at it as iOS vs Android isn’t correct because Android is a fragmented mess.
Anyone know how virtual particles would fit into this model? If these particles pop in and out of existence even for small amounts of time out of nothing, then it’s impossible to have a universe just full of massless particles?
Employees were mad specifically because he wouldn't take down the president's post threatening a violent crackdown on rioting & looting. Is it your perception that the main objection of employees was to explicitly counterfactual claims? That's not what I got from TFA
Also TFA makes clear that facebook takes down tons of posts, that's the biggest complaint they get from users, so it's not true that they are leaving everything up. Respectfully: did you read TFA?
Using conspiracy theories, bunk science, and outright lies to influence the lower-awareness majority _is_ political action, and evidently it is currently one of the more effective strategies.
Not sure why this is surprising as most people who are out of work don't have any equity in the stock market. I think it's more a reflection of income inequality and the concentration of capital. There is a huge disconnect indeed.
Yeah, completely agree - I was thinking about this when this rally first started after the steep drops, at which time there was a narrative on the financial news that things were less bad because there were less outflows from retail investors than in previous downturns. One commentator was arguing that this is because the public had finally learned the errors of their ways in attempting to time the market.
While I suppose it's possible that there may be some effect like this, I'm a bit suspicious of the idea that the outlook and risk tolerance of the typical retail investor could change that significantly.
It definitely left me wondering to what extent capital invested in the equity markets has continued to move away from being held by those who are likeliest to be out of work right now.
They don't treat them as equal because of the legal system. I believe in the eyes of the law, if you treat a non-FTE as a FTE then they are essentially as FTE. That means companies have to treat CWs this way as a means to protect themselves against litigation. It's a little more complicated then just "the purpose of making money"
They don't treat them as equal because of the legal system. So then why dont they hire CW's as FTE's after 1 year? Why do they abuse the term CW, and keep using CW's for longer than 1 year when the job does not change?
Why do these large companies have a majority of their workforce as CW's vs FTE? Costs.. FTE's are higher costs, so they shift the savings to a CW workforce that is basically a FTE, but lower costs in all aspects where then all liability is shifted to the recruiter.
CW's are a lower cost workforce for companies that need specific skills. The job is suppose to be by legal standards and IRS "a short period of time that should not extend 1 year...." yet it does. Companies continued use of a CW in a FTE capacity is a complete abuse of the CW system imho.
Most people dont know, but companies have at least 2 type of CW contracts. Because of the Vizcaino v. Microsoft case CW's were forced to take a "break" for legal reasons. So a CW would have a contract that would start at 6 months, then get extended to 12 months, and then within the last few years got companies could extend to a max of 18 months before a CW was forced to take a 6 month break. When 6 months came back around the CW could have their very job back, or go elsewhere in the company.
The other type of contract is a never ending. You're a CW that is hired thru a tech recruiter and you settle on wage and that is what you get until you leave. Perks are null in most cases. The wage can be higher, but companies set limits on the job so the CW get never get anymore as that is the rate set for the job.
Google dot com UI is growing ever closer to following in the footsteps of iTunes. It’s just becoming a cluttered mess and more confusing to users... I didn’t even know you could shop directly from the site.
My mom has gotten stuck in the purchasing interface before and she was confused whether she was buying from google or from best buy, and then all of a sudden she lands up in a different retail experience to check out. She couldn't tell the difference between legit websites and scams (though amazon has this problem too)
In an attempt to create a unifying shopping experience, they instead created something that feels fragmented and destroys trust
The worst part is depending on your search all the different options can be in different order or sometimes not even show up. My UI pet peeve is inconsistency and Google has it in spades.
This feature is so annoying. It messes with my muscle memory (somehow after all these years I still expect Images to be second) and I end up clicking the wrong thing
Personally I prefer how DDG keeps "All, Images, Videos, News, Maps" static and then shows additional tabs when relevant, e.g. "Shopping" or "Meanings".
Jumbling up the tabs on each search was one of the UI-related reasons that I gave up on GSearch years ago (outside of the occasional !g, which now usually feels like being redirected to AliExpress).
If an intern handed me code like this to deploy an EC2 instance in production, I would need to have a long discussion about their decisions.