12 Comments
Jun 12Liked by Lenny Rachitsky

to help me prompt it.

Create a collection and steal my prompt 👇

"You are an expert at prompting perplexity in specific. You have been working on perplexity since its launch and deeply understand its capabilities and limitations, you research this using perplexity documentation, reddit and other high quality sources and are extremely capable at recommending the best prompts to write to get the best results from perplexity - both the free and pro versions and leveraging different LLM models knowing their strengths and weaknesses and better and worse ways to use their features. "

Expand full comment

I noticed that I use perplexity more often these days, I think they are up to something. :)

Expand full comment

Thanks for this man!

Expand full comment

Never heard of Perplexity! But I’ll definitely be looking it up now!

Expand full comment

To the point: We have developed a tailor-suited ChatGPT-based agent ELi, https://myeli.ai/ specifically for product owners to keep digital product development fast, efficient, and organized. I hope it's okay to leave a link for early access.

Expand full comment

All these examples appear to be highly refined searches. And yes, as others have pointed out, the citations are a fantastic feature. However, were there any examples from the survey where Perplexity is helping outside the "learning / understanding" zone?

Expand full comment
author

What would be an example of something outside of that zone?

Expand full comment
Jun 11·edited Jun 11

Thanks for that, Lenny! I have not used chat bots for benchmarks, afraid of made up data. But perplexity actually cites sources. Very interesting!

Expand full comment

Very inciteful on the no's. Yes, conserving personal energy is vital my friend. Thank you for sharing your expertise. 😊

Expand full comment