By using this site, you agree to our Privacy Policy and our Terms of Use. Close
ConciousMan said:
IcaroRibeiro said:

AI usage is completely fine to things like coding and bug fixing

In my experience Claude is able to produce much higher quality code than most (if not almost all) developers, because the model is trained with the best coding standards from public repositories, standards close to impossible to be religiously applied by all members of a team for production-level coding

Being against AI in software development is like being against using machines in agriculture or in manufacturing

Ok I don't know what kind of experience you have, but since when software development is just coding? The way more important thing is to make things fixable for me. Does Claude Code make you understand the code more or less? For me coding is being less important, making things scalable and secure is what brings money IMO.

Claude makes significantly more readable code than humans, it even comes with pre-generated docstring and comments, which developers HATE to write 

The rest of your post is attacking a strawman. I talked specifically about coding, and then you come with architecture, security and system design

AI can be used to help with those as well, but even if you put an agent on meetings to analyize technical discussions it can't plan, model and design a software by itself. That's why agents still need specifications, documents and human supervision 

For coding itself though? I'm sorry but Gen AI is already ahead of humans