Prompts, conversations and framing problems

When LLMs were first released, it was more important to be a 'prompt engineer', to include all details in a single prompt.

Now, using LLMs (through tools like ChatGPT) is more 'conversational' and we often write lots of follow-up prompts, like: “Explain it simpler”, “What about …?” or “Can you expand on …?”

However, it's still important to think before you prompt, about what details you can include, and the problem that you are trying to solve. This is called framing the problem.