So it has only 252GB of actual ”AI” memory making it “useless”/toy for actual real world AI workloads(I.e it can’t replace something like opus 5)
945 karma · joined November 17, 2021
So it has only 252GB of actual ”AI” memory making it “useless”/toy for actual real world AI workloads(I.e it can’t replace something like opus 5)
Can we pause the AI development after the AI slop is “fixed” perhaps with something less than 10.000 agents?
And I think this is really the best of both worlds because AI coding works best when you give it constrains and without my toolchain the AI is generating crap code I wouldn’t ever want to read but using the libs I wrote and know I can constrain it to write decent code that works “at scale” . The more opinionated the tools/libraries the better outcome you can expect.
Another great usage of AI is for refactoring. Before AI you could have a great idea how to make the library better/more ergonomic but the sheer amount of refactoring would make you just eat the debt and basically live with your own “slop”. Now with AI you can produce the reference design for the refactoring (I.e let’s make all these functions actually methods) and the job is done in minutes/hours. This inevitably leads to better designed software.
We don’t care if Trump is doing a deal with criminals to steal oil from Venezuela either so why should the world be so eager to help him steal oil from Iran as well?(and perhaps do a deal with the meanest of them all)
Next article should be about UK’s dependence on U.S for their nuclear deterrence. Someone give them that tip
I am at version 4 and I have a “open in vscode” button on each task/orb/instance just in case the UI is not enough and you want to corner the agent really close
In the long term the world will adjust to the new reality. As a simple example take the oil crisis from 1970s. The U.S adjusted and now is the world’s biggest producer.
It will take time but in 30 years I would be surprised if Europe would still buy U.S weapons, let U.S tech services rule EU digital markets and so on.
I think the lack of syntax highlighting was one reason for that. It makes you think more about the code you write and how it should compose, while a fully fledged IDE encourages you to just throw more code at the problem.
I think the ACME way should be used for love of code and ideally when you want to create libraries that stand the test of time. When you just want to get things done fast bring the full IDE and now some LLM vibes and it’s done.