The motivation for building simpleaichat was indeed a direct reaction to the frustrations of using LangChain, with a Hacker News thread full of similar complaints being the primary catalyst for development: https://news.ycombinator.com/item?id=35820931
This package isn't trying to ride the AI hype wagon for venture capital as often said on AI submissions on HN: it's to fill an actual demand, and one I personally needed even if no one else uses simpleaichat.
There's still a lot of work that needs to be done with the package (it's missing important demos such as working with embedding vectors, which is a separate project I have in mind born out of annoyance) but I'll be putting forth the time on it.
This package isn't trying to ride the AI hype wagon for venture capital as often said on AI submissions on HN: it's to fill an actual demand, and one I personally needed even if no one else uses simpleaichat.
There's still a lot of work that needs to be done with the package (it's missing important demos such as working with embedding vectors, which is a separate project I have in mind born out of annoyance) but I'll be putting forth the time on it.