Okay but 99% of corporate codebases are already insane crap. The LLMs write boring, overly safe code, but at least they don’t make obvious IDOR bugs like my human coworkers.
Yeah but when you’re gone nobody is going to even be able to tell the difference between one set of code or the other. Instead the competent programmers will be replaced with tech priests preaching to the machine.
Embracing this technology eliminates competency. Consider how the art industry is currently refusing to embrace it and instead fighting it, what does this do? It ensures that the future has artists who remain competent and does not turn the field of art into a bunch of people who give prompts to an AI.
To really understand this you need to imagine a future where the art field embraced AI and all manual art stops, instead being replaced by people that give prompts to AI to make art. This is the future of the programming field by embracing AI. All future programmers when the manual programmers phase out will just be people that only know how to prompt.
It is incredibly shortsighted and only the realisation that everyone in the field will become like ai “artists” once the current experience phases out really makes this clear. It will suffer a total skill collapse.
But the art field is making final products that have no formal correctness. The programming field is making infrastructure that never gets seen by end users, and we do have formal verification. Two completely different things.
The skill collapse already happened after the 90s which is why all the software is so shit right now. Now the competent people can take over again and rewrite everything in immutable functional languages with complete unit test coverage and viral copyleft licenses.
I work with frontier models every day that I participate in gainful employment activities, and I don’t think I would call them reliable. Definitely not reliable enough to design complicated software correctly on a language there’s relatively little training data on.
They struggle with design because they can’t tell a good idea from a bad idea without writing a comparative essay first. You can still make it work as long as no more than a third of your sessions are about implementation. Everything else should be about exploration and planning. The human drives the design.
It’s an equal amount of effort but 10 to 100 times more productive so I actually have time to add unit tests and error recovery strategies before the managers come asking questions.
I guess if your company isn’t totally dysfunctional it’s not as helpful but I’ve never even heard of a corporate dev environment not ran by idiot PMs
Okay but 99% of corporate codebases are already insane crap. The LLMs write boring, overly safe code, but at least they don’t make obvious IDOR bugs like my human coworkers.
I’ve seen LLMs put sensitive secrets in git just like any incompetent junior dev
Not to mention that it’s quite common to see them delete the database if they can get access to it
critical support?
It’s done randomly and infrequently so at best it’s a form of adventurism if you’re ok with anthropomorphizing.
Harness + got hooks + formal verification trivially prevents everything like that. You’re just holding it wrong.
Yeah but when you’re gone nobody is going to even be able to tell the difference between one set of code or the other. Instead the competent programmers will be replaced with tech priests preaching to the machine.
Embracing this technology eliminates competency. Consider how the art industry is currently refusing to embrace it and instead fighting it, what does this do? It ensures that the future has artists who remain competent and does not turn the field of art into a bunch of people who give prompts to an AI.
To really understand this you need to imagine a future where the art field embraced AI and all manual art stops, instead being replaced by people that give prompts to AI to make art. This is the future of the programming field by embracing AI. All future programmers when the manual programmers phase out will just be people that only know how to prompt.
It is incredibly shortsighted and only the realisation that everyone in the field will become like ai “artists” once the current experience phases out really makes this clear. It will suffer a total skill collapse.
But the art field is making final products that have no formal correctness. The programming field is making infrastructure that never gets seen by end users, and we do have formal verification. Two completely different things.
The skill collapse already happened after the 90s which is why all the software is so shit right now. Now the competent people can take over again and rewrite everything in immutable functional languages with complete unit test coverage and viral copyleft licenses.
Great, so why has that not happened?
Cause the models have only been reliable for like 6 months, and only cheap for the past two weeks
I work with frontier models every day that I participate in gainful employment activities, and I don’t think I would call them reliable. Definitely not reliable enough to design complicated software correctly on a language there’s relatively little training data on.
They struggle with design because they can’t tell a good idea from a bad idea without writing a comparative essay first. You can still make it work as long as no more than a third of your sessions are about implementation. Everything else should be about exploration and planning. The human drives the design.
To be honest with you, that sounds like more effort than just making it yourself for the most part.
It’s an equal amount of effort but 10 to 100 times more productive so I actually have time to add unit tests and error recovery strategies before the managers come asking questions.
I guess if your company isn’t totally dysfunctional it’s not as helpful but I’ve never even heard of a corporate dev environment not ran by idiot PMs
Cool! Where is the git repo?
Im still working on it but you can use Pharo for now