Thread by @ItakGol

The LLM tip of the day #1-

I have discovered a small yet crucial tip for producing superior code with GPT-4, which can help you increase productivity, deliver faster results, and enhance accuracy.

#LLMTipOfTheDay

>>>Thread>>>

When you are asking an LLM, let's say GPT-4, to code something, it is fair to say (although I am simplifying things a bit) that it is converging to the expectancy level of the training data.

What do I mean?

>>>

It was trained on code from less skilled programmers (many) as much as it was trained on code from rockstars programmers (less). So it is generating code intuitively that is an average of what it has learned.

But you want highly skilled code, right? So just ask for it!

>>>

In this example, I'm asking for it in the System section, as if I am Jeff Dean, a top-world expert in Coding and Algorithm Design at Google.

Surprisingly, or not?, in 75% of the examples I have checked, the code was much better!

>>>

Obviously, this paradigm can be utilized in other domains, from poetry to product marketing.