“Understands intent well and hammers until it gets there. “
If there’s anything I learned over the past 12-18 months is that this is a recipe for disaster, except for throwaway stuff.
I thought most senior engineers settled on the fact that steering a model yields much better results?
It's very possible that would be the best strategy over the last 12-18 months and now that this is released it is no longer the best strategy.
I wouldn’t call it a recipe for disaster, but oh boy if you leave an agent that “hammers until it gets there” on its own with an underlying bug in a dependency…