I was “there” with Claude as you describe about 3 months ago. Since then, Claude has stepped up to being able to create fully functional microservices. It helps if you completely specify what you want, it helps if you don’t specify funky libraries or other tech that has poor support on the internet, it helps if your total ask amounts to 1000 lines of code or less - but I have gotten up around 3000 lines before Sonnet 4 choked a few times.
Before this, my AI queries were mostly limited to specific API function call syntax, and they would only be right about 2/3 of the time, which beats randomly trying things myself until I eventually guess the right variation… Yes, it’s better to consult the documentation - when it’s available - it’s not always available.
yea so maybe the resulting future is that the tools can only work with really popular libraries that have lots of people talking about them on stack overflow in the year 2024 or whatever, and new smaller potentially interesting libraries will have a harder time seeing adoption
maybe the resulting future is that the tools can only work with really popular libraries that have lots of people talking about them on stack overflow in the year 2024 or whatever, and new smaller potentially interesting libraries will have a harder time seeing adoption
Yeah, that’s the future I’ve been living since about 2005. The alternative to letting the world be your support desk via stack overflow and similar is to develop killer examples and API documentation for your own libraries so the AI (and everyone else) can learn from that. Qt was a great example of this starting in the early 2000s.
The dark future is where you have competitors “poisoning the well” spreading false information about your tech in the normally reliable channels, then having AI amplify that for them. This, too, is already happening to some extent - more in the political sphere than the technical space, but it’s everywhere to some extent.
I was “there” with Claude as you describe about 3 months ago. Since then, Claude has stepped up to being able to create fully functional microservices. It helps if you completely specify what you want, it helps if you don’t specify funky libraries or other tech that has poor support on the internet, it helps if your total ask amounts to 1000 lines of code or less - but I have gotten up around 3000 lines before Sonnet 4 choked a few times.
Before this, my AI queries were mostly limited to specific API function call syntax, and they would only be right about 2/3 of the time, which beats randomly trying things myself until I eventually guess the right variation… Yes, it’s better to consult the documentation - when it’s available - it’s not always available.
yea so maybe the resulting future is that the tools can only work with really popular libraries that have lots of people talking about them on stack overflow in the year 2024 or whatever, and new smaller potentially interesting libraries will have a harder time seeing adoption
Yeah, that’s the future I’ve been living since about 2005. The alternative to letting the world be your support desk via stack overflow and similar is to develop killer examples and API documentation for your own libraries so the AI (and everyone else) can learn from that. Qt was a great example of this starting in the early 2000s.
The dark future is where you have competitors “poisoning the well” spreading false information about your tech in the normally reliable channels, then having AI amplify that for them. This, too, is already happening to some extent - more in the political sphere than the technical space, but it’s everywhere to some extent.