@Raynerdtech I don't believe to you, I fact check and before doing that, chatgpt won't send you to a competitor and it was built around safety and it loves explanations.
something interesting happened while studying robotics today.
i understood what a coordinate transformation problem was asking. i could follow the explanation. i could look at the solution and say, yeah, that makes sense.
then i closed the solution and tried to derive it myself.
nothing.
and i think this is one of the biggest traps in technical learning: recognition feels dangerously similar to understanding.
when someone explains rotation matrices, coordinate frames, jacobians or control theory cleanly enough, your brain follows the logic and gives you the feeling that you know it. but following someone elseβs chain of reasoning is not the same thing as being able to generate that chain yourself.
so iβm changing how i study robotics.
less consuming explanations. more sitting with stupidly small problems until i can formulate the solution from first principles. draw the frames. reason about which direction each axis points. predict the answer. derive it by hand. then write the code.
LLMs make it incredibly easy to skip this discomfort. you can build surprisingly complex systems while understanding very little of what is happening underneath.
i donβt want that.
iβd rather spend an hour struggling with a basic coordinate transformation today than build an impressive robot tomorrow whose mathematics i cannot explain on a whiteboard.
the struggle is probably the curriculum.
@oprydai Bro definitely knows what he's doing,
I usually face the same problem especially when learning new languages or logic in a particular language. So I calm down break it into pieces and discover where the fuss actually originated from!
2 years ago as a developer, I was earning β¦150k/month.
Got an opportunity to move to the UAE. π¦πͺ
Today, I picked up my first dream BMW. π€
Thank you, JavaScript & Java.