I'm hacking on a library (https://github.com/gsuuon/ad-llama https://github.com/gsuuon/ad-llama) inspired by guidance, but in TS and for the browser. I think st

I'm hacking on a library (https://github.com/gsuuon/ad-llama https://github.com/gsuuon/ad-llama) inspired by guidance, but in TS and for the browser. I think structured inference and controlled sampling are really good ways of getting consistent responses out of LLM's. It lets smaller models really punch above their weight. I wonder what other folks are building on this sort of workflow? I've been playing around with it and trying to figure out interesting applications that weren't possible before.