[mandliya]

I am curious if you tried something like langchain[1] it would solve the problem of not remembering previous conversations. In one of the examples people have used an entire company notion database on top of GPT-3 to answer questions specifically from that database.

PS: not tried myself.

[1] https://github.com/hwchase17/langchain

[swyx]

if i understand correctly it doesnt really remember all previous conversations, it summarizes them to pack more into the limited context window right? thats kind of just kicking the can down the road vs solving the problem of actual long term memory

[Xeophon]

Isn’t this the same as replying to ChatGPT in the same conversation?

[frabcus]

It more let’s the bot can query a database (using a DSL you teach it) when it wants to.

So for example if it couldn’t remember an anecdote from the clients past, I guess it could search the history?

eg ask to search for “Sarah" and then use all stories about her as part of a prompt and iteratively rerun itself.

I think humans would still beat it at synthesising patterns drawn from disparate sessions. But it sounds doable to code…