As you know the capacity of modern LLMs supports to creation of mostly succeed Python codes. What happens if we use this capacity in the real coding future? Just write a prompt and explain what you want and what you expect. It just generates the codes and runs and exports the output. It is genius or not ? Speed, cost against efficiency