3 ms·
> As of today, none of the existing models can meaningfully handle mid-size tasks on five services with 10k+ LOC each My FAANG's codebase is a few orders of ma
by solenoid0937 4mo ago
> As of today, none of the existing models can meaningfully handle mid-size tasks on five services with 10k+ LOC each
My FAANG's codebase is a few orders of magnitude larger and agents do an excellent job of handling mid sized tasks completely autonomously.