How Should We Use Claude Code?
This is my sorta complex view on using Claude Code
Published: April 6th, 2026
If you don't know or wasn't able to tell I create a lot of projects.
However a lot of people I know try to automate production with the
super smart Claude Code, the smartest AI mind. Unlike others I always
try and keep my code purely written by humans due to a sense of pride
for being able to create something from scratch and also being able to
solve a puzzle myself. Both views have some issues so I've come to a
stance towards Claude Code that I'd like to share.
As I hinted at earlier, there seem to be two opposite views / stances
on AI [Claude Code especially]: either very much in favor of or very
much against the use of AI. The problem with these extremes is that
there isn't a good reason to fall into either because they both cut
out human potential for new innovations. When computers were first
invented people didn't understand them and it led to people
distrusting technology but then computers became a vital part of
society. Don't believe me? My estimate is that at least 5 different
computers [including the one I programmed everything involving this
website on] that transmitted this data to your device. AI is just like
that.
From how I see things both extremes must be wrong then. If AI is the
best tool that people can use to develop innovations, we should use it
rightly. That doesn't always mean creating stuff for you but it can
mean debugging stuff for you. Not using it in a world where people
won't stop to listen to why you wouldn't is foolish.