Finding the fault
The first time our robot failed at regionals, I stayed after everyone else had left to figure out why.
It wasn’t the code — it was a loose servo mount I had rushed the night before.1
That evening in the empty workshop taught me more about engineering than any win since.2