Comment Comment by root_axis We've repeatedly seen that these test-driven LLM rewrites consistently produce absolute garbage. parent root_axis 4 months ago
