So we're basically prompt engineering... for our prompt engineering? Feels a bit like 'Inception,' lol.
This is well written so if AI wrote some of it I can't tell, but there is an advantage in not using AI at all to write. It makes you think better, and carefully craft sentences. It is also more respectful to the audience.
Stop calling it engineering.
Let's say I write a prompt and instruct ChatGPT to analyze the prompt and apply the best guidelines on prompt engineering to it to generate a new prompt that does the job better. Then I take that generated prompt, give it to DeepSeek with the same instructions, and take the output, give it to Claude, and so on through all the released LLMs until we circle back to ChatGPT. This is something like the telephone game, but will the final result be a better or worse prompt than the original?
Garbage in, garbage out - GIGO - still seems to apply to LLMs. It might be nice if LLMs would respond with 'this prompt doesn't compile, try again' while emitting an error report, like compilers do, if some minimal standard wasn't met.
I'm on hot-take fire today, so may as well keep it going. I'm of the opinion this is the new field of programming. Think of it like Game Engine scripting. The engine is made, but all the level design is going to have to be done with scripting. People that know how to design innovative and practical prompt solutions are going to be quite valuable. In fact, Prompt Engineering is all that's left. Don't even try to out-hot-take me on this one.
Great, now we have a service for prompting prompts for a prompter, so prompt guessers don't have to do the guessing themselves any more. I'm sure there will also be hallucinated prompts in there. I'm curious whether a self-reinforcing prompt hallucination loop could emerge.
"Complete Guide to Meta Prompting while recommending our Product" would be a more honest title.
I personally reject advice that is muddled with directly offering their own services: Its conflict of interest in my face.