@IterIntellectus Agreed.
LLMs act as a coefficient, but if the variable theyβre multiplying (knowledge/ability/etc of the person using it) is less than 1, the result can shrink instead of grow.
@michaelybecker I've frequently used 'smoke test' as a phrase well before LLMs.
Usually when making a change to some underlying code/logic that shouldn't cause an issue, but I want to do high-level testing (smoke test) of the consuming code or functionality to make sure we didn't break anything.
@wesbos I agree 6 db calls are negligible in the grand scheme of things, but when using a framework do we sign away our right to ask questions about what's going on under the hood?
Was he 'sweating' 6 calls or was he trying to figure out the 'why' of the 6 calls?