3 ms·
Setting aside the AI angle, perhaps just recording the assumptions in a way that can be measured would be enough. Tooling, runtime sampling, or just code revie
by jeffdavis 6y ago
Setting aside the AI angle, perhaps just recording the assumptions in a way that can be measured would be enough.
Tooling, runtime sampling, or just code review could reveal when the assumptions go awry.