The consensus among commenters is that while LLMs have great capacities for handling repetitive tasks, they should not be relied upon for decision-making or complex logic implementations. Instead, LLMs should serve as auxiliary tools for tasks where they can shine, such as executing known processes and improving workflows. For effective use, LLMs should be embedded in broader systems orchestrated by traditional engineering practices, rather than replacing them entirely. This approach includes utilizing small LLMs for faster responses and ensuring user experiences avoid frustrating interfaces that rely excessively on text input.