Evidence receipt / belief
Published · transcript-backedShawn Wang: belief
10 Jan 2025 Latent Space Beating Google at Search with Neural PageRank and $5M of H200s — with Will Bryk of Exa.ai
“Basically, I think dating, uh, research, like I just wanted to like spell out more things, like just the big verticals.”
Source trail
Everything needed to verify it.
- Speaker
- Shawn Wang
- Attribution
- Verified speaker
- Claim type
- belief
- Recorded
- 10 Jan 2025
- Publisher
- Latent Space
Transcript context
…Yeah, sure. So, I mean, there are so many applications of XR and we have, you know, many, many companies using us for very diverse range of use cases, but I'll just highlight some interesting ones. Like one customer, a big customer is using us to, um, basically build like a, a writing assistant for students who want to write, uh, research papers. And basically like XR will search for, uh, like a list of research papers related to what the student is writing. And then this product has. Has like an LLM that like summarizes the papers to basically it's like a next word prediction, but in, uh, you know, prompted by like, you know, 20 research papers that X has returned. It's like literally just doing their homework for them. Yeah. Yeah. the key point is like, it's, it's, uh, you know, it's, it's, you know, research is, is a really hard thing to do and you need like high quality content as input. Oh, so we've had illicit on the podcast. I think it's pretty similar. Uh, they, they do focus pretty much on just, just research papers and, and that research. Basically, I think dating, uh, research, like I just wanted to like spell out more things, like just the big verticals. Yeah, yeah, no, I mean, there, there are so many use cases. So finance we talked about, yeah. I mean, one big vertical is just finding a list of companies, uh, so it's useful for VCs, like you said, who want to find like a list of competitors to a specific company they're investigating or just a list of companies in some field. Like, uh, there was one VC that told me that him and his team, like we're using XR for like eight hours straight. Like, like that. For many days on end, just like, like, uh, doing like lots of different queries of different types, like, oh, like all the companies in AI for law or, uh, all the companies for AI for, uh, construction and just like getting lists of things because you just can't find this information with, with traditional search engines. And then, you know, finding companies is also useful for, for selling. If you want to find, you know, like if we want to find a list of, uh, writing assistants to sell to, then we can just, we just use XR ourselves to find that is actually how we found a lot of our customers. Ooh, you can find your own customers using XR. Oh my God. I, in the spirit of. Uh, using XR to bolster XR, like recruiting is really helpful. It is really great use case of XR, um, because we can just get like a list of, you know, people who thought about search and just get like a long list and then, you know, reach out to those people.…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.