
YB
6.1K posts



One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.


@benspringwater the desire to hold thoughts in our hands is as old as humanity







most startups treat their launch as the story. it’s definitively not.. it should be just the opening line. what actually happens now is that there is typically one big launch video, then silence, then occasional product announcements that nobody asked for & nobody remembers. there’s no connective tissue. each piece of content has to justify itself from zero because there’s no accumulated world for it to inhabit. what you should be doing is designing the launch as chapter one of a long book… i.e. don’t fucking create a standalone film. every week you’re adding a page to that damn book.. this is a feature, a use case, a small revelation about what the product knows & why that matters. the content continues. what does this do? well this compounds like hell. audiences who follow a narrative become invested in its resolution. they feel the absence when a chapter’s missing. that’s a completely different relationship than “i saw their launch video once.” this is what we will do with skye.















