Re:Show HN: I built a LLM-powered Ask HN: like Perplexity, b...
by Valk3_ on Sunday May 19, 2024 @08:14PM
(#40410784)
Awesome, looking forward to it!
Re:Show HN: I built a LLM-powered Ask HN: like Perplexity, b...
by jnnnthnn on Wednesday May 22, 2024 @11:19AM
(#40442092)
Blog post is now live:
- https://news.ycombinator.com/item?id=40442039
- https://jnnnthnn.com/how-to-build-your-own-perplexity-for-an...
Yes, the underlying dataset very much conditions the quality of the responses. Additionally, the retrieval strategy is also a really important factor (and that is something which I haven't had time to extensively optimize).
I'm writing a blog post that will answer your questions! Will post it here when it's fully baked.