Yeah experimentation is probably the best use case of LLMs
And at the end of the experiment I usually have my solution
And at the end of the experiment I usually have my solution