Kriptering 2 hours ago
I think we've gotten far enough with LLMs where we dont need to prompt any more for leads (connected a bunch of apis for enrichment and signals); leads are correlated with a context brain I made based on latour's actor - network theory
maybe too heady of an approch, but w/e it works; and the podcasts are fun to listen to