Fun challenge for slop-curious and slop-native programmers: race the LLM to fix the issue or implement the feature. So far My intuition helps me use fewer "operations" and debug rounds even though the LLM can read and spit code faster.