Evidence receipt / evaluation
Published · transcript-backedAravind Srinivas: evaluation
19 Jun 2024 Lex Fridman Podcast #434 – Aravind Srinivas: Perplexity CEO on Future of AI, Search & the Internet
“You can do that too when you’re small just to delight users, but it doesn’t scale.”
Source trail
Everything needed to verify it.
- Speaker
- Aravind Srinivas
- Attribution
- Verified speaker
- Claim type
- evaluation
- Recorded
- 19 Jun 2024
- Publisher
- Lex Fridman Podcast
Transcript context
…So constantly you come up with an issue with a particular set of documents and particular kinds of questions that users ask, and the system, Perplexity, it doesn’t work well for that. And you’re like, ” Okay, how can we make it work well for that?” Correct, but not in a per-query basis. You can do that too when you’re small just to delight users, but it doesn’t scale. At the scale of queries you handle, as you keep going in a logarithmic dimension, you go from 10,000 queries a day to 100,000 to a million to 10 million, you’re going to encounter more mistakes, so you want to identify fixes that address things at a bigger scale. Hey, you want to find cases that are representative of a larger set of mistakes.…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.