Yeah the over optimize before getting the proof of concept/getting it working minimally should always be recognized as gambling, as the n you may lose all that

Yeah the over optimize before getting the proof of concept/getting it working minimally should always be recognized as gambling, as the n you may lose all that optimization if the stuff doesn’t work with it down the line