2 ms·

> "running shoes I looked at last week" / "Here's what I found in your browsing history:" Does that need an LLM at all?

• smsm42 18d ago
  Not really, but it's probably easier to make it on top of LLM than to make specially-purposed tool for it, if we talking in terms of time-to-market effort.
  • flatline 18d ago
    SQLite FTS could have done this a decade ago. We've had good local search capabilities for two decades and they either been underused or abandoned (e.g. Google Desktop, Yahoo! Search). This may be a reasonable projection of where AI is headed. You can do a lot locally but there is too much incentive to centralize around cloud infrastructure, then the privacy concerns make that prohibitive and we end up with what feels like a false choice of cloud or bust. Maybe it's LLM hype that will bring more powerful capabilities to the desktop?
    • smsm42 14d ago
      I think classification and clusterisation is more important than search per se. "Where did I see that article about that weird psychological effect where people remember things they haven't seen" is not resolved by direct search, but can be - with some work - helped by language models and NLP. One doesn't need full blown frontier LLM for that, but bare word indexing would probably not do either.