articles like these make me think that coding with AI is a little bit like writing Perl code: if you know what you’re doing, you can do brilliant things very quickly, but if you don’t, you can make spaghetti very quickly.
That's a great analogy and is something I experience every second day. Once a week I do a full second pass of a manual review on the generate AI code. Very often I find myself in a situation were I do not really understand the recently AI generated code anymore or find it hard to read, so I either rewrite it manually or tell the LLM to make it more readable. And this is just one part. If you really would like to get a long-term maintainable software product, AI code suddenly isn't that much of a speed boost anymore. Maybe a little bit, but the initial wow effect is very ephemeral.
That's a great analogy and is something I experience every second day. Once a week I do a full second pass of a manual review on the generate AI code. Very often I find myself in a situation were I do not really understand the recently AI generated code anymore or find it hard to read, so I either rewrite it manually or tell the LLM to make it more readable. And this is just one part. If you really would like to get a long-term maintainable software product, AI code suddenly isn't that much of a speed boost anymore. Maybe a little bit, but the initial wow effect is very ephemeral.