{"type":"rich","version":"1.0","title":"Dustin wrote","author_name":"Dustin (npub1mg…gpdjc)","author_url":"https://yabu.me/npub1mgvwnpsqgrem7jfcwm7pdvdfz2h95mm04r23t8pau2uzxwsdnpgs0gpdjc","provider_name":"njump","provider_url":"https://yabu.me","html":"If you didn't see the full slide deck, the link is here:\n\nhttps://dtdannen.github.io/Coordinating_(DV)Machines_9JAN2023.pdf\n\n(Warning - it’s a lot of content mashed together)\n\nGenerally, you can think of DVMs as being the “actions” or “tools” or “functions” that an agent would call. (These are terms other projects are using; OpenAI refers to them as tools and functions)\n\nOnce we have a multi-step DVM chain (like a DVM passes it’s output to another DVM before the final response is given back to the user) then we have an equivalent to a hardcoded LangChain agent. \n\nMore flexible agents will choose dynamically which DVMs to use, in which order, to solve problems. That’s where things will get very interesting."}