I think this person just learned about Vibe coding and is in the "Peak of inflated expectations" [0] as evidenced by this title: "» you can understand anything, now".
Understanding is not just doing. Understanding is being able to build something up from first principles. The author of this post will better understand the difference when he hits a non-trivial bug or the project grows past a certain size.
I have 40 years of programming experience, started with assembler, nowadays mainly Go, K8s and the whole enterprise shebang. I’m a big fan and supporter of TDD and XP.
Claude Code will change your life when you learn how to program with it. However, if you are a programmer with not a lot of desire for automated tests and specs/designs, you are probably not going to be successful with it.
The art of coding has become a commodity. Validation and verification are the new art.
> [...] will change your lice when you learn how to program with it.
But how (honest question) do you learn hoe to peogram with it. All I see is people using it to program and stop thinking about all the steps required between start and goal. That's not learning, that's assisted doing and that will only get you as far as the tooling (the assistent) goes. I've yet so see someone that learned coding with AI and was then able to do the same job without it.
> Claude is so far ahead of manual search that it’s not a comparison.
This is frustratingly true, but I don't think Claude is ahead of what search used to be. We have forgotten that google used to be considered magic back when it worked.
LLM tooling will be no different, it'll be magic whilst the money flows then get suddenly terrible.
how ?
I think this person just learned about Vibe coding and is in the "Peak of inflated expectations" [0] as evidenced by this title: "» you can understand anything, now".
Understanding is not just doing. Understanding is being able to build something up from first principles. The author of this post will better understand the difference when he hits a non-trivial bug or the project grows past a certain size.
[0] https://en.wikipedia.org/wiki/Gartner_hype_cycle
Fixing a non-trivial bug is a great way to learn - assuming they don’t give up.
By virtue of being generators subtly broken stuff, LLMs are well positioned to create very nice learning material.
Same thing about growing the project - having to deal with something too big for AI is a very valuable experience.
And, in my experience, some of the purely human made codebases are strictly worse than LLM-made :-)
Isn’t that how a lot of us learned — buy typing the code out of back of a magazine? Then spending hours trying to debug a typo somewhere.
I didn’t realize how close LLMs are to the old magazines. Let it give you a seed, then use that springboard to learn everything else.
I have 40 years of programming experience, started with assembler, nowadays mainly Go, K8s and the whole enterprise shebang. I’m a big fan and supporter of TDD and XP.
Claude Code will change your life when you learn how to program with it. However, if you are a programmer with not a lot of desire for automated tests and specs/designs, you are probably not going to be successful with it.
The art of coding has become a commodity. Validation and verification are the new art.
> [...] will change your lice when you learn how to program with it.
But how (honest question) do you learn hoe to peogram with it. All I see is people using it to program and stop thinking about all the steps required between start and goal. That's not learning, that's assisted doing and that will only get you as far as the tooling (the assistent) goes. I've yet so see someone that learned coding with AI and was then able to do the same job without it.
with all these hypes around LLM my first thought were using it for writing automated test and rspecs
[dead]
[dead]
the way this guy writes code different than the way I writes code
> Claude is so far ahead of manual search that it’s not a comparison.
This is frustratingly true, but I don't think Claude is ahead of what search used to be. We have forgotten that google used to be considered magic back when it worked.
LLM tooling will be no different, it'll be magic whilst the money flows then get suddenly terrible.