Look I was taught and told that optimisation should be done at the end of development because if you refracter code to optimise it then there’s a chance that breaks something else and after continuing development the code you ‘optimised’ will end up being refractored again anyway (and probably end up less optimised).