The speaker says that investing more time in the initial prompt or providing better context can prevent iterative rework that uses more model capacity.

“oh like it did a lot of work and you're like oh i don't like this like can you like undo this and redo it and and then you're like iterating on this like thing that the model could have done if you had like spent more upfront time or given it better context you know and instead it's sort of like you're like nope don't like that design try this or like you mess this up or something like that and then that just eats up so much more of like you know your usage”

“and so that's like i think maybe like a key like tip both for like efficiency as well right and yeah i think like context and not just like context on like”