As a primer, I stitched together a simple ChatGPT-based app that queries my Notion database of design resources and links I've been bookmarking over the past few years.

Took me part of an afternoon.

I connected my Notion database with LlamaIndex and OpenAI, fronted with a simple React app (all from an example repo, I'm not superhuman). Literally zero to hero in a few hours.

It costs a little too much in OpenAi API calls to release publicly and would require some design work that I'm not willing to put in right now.

But it was a great lesson in how cheap and accessible these tools have become -- and we're barely out of the gate.

I don't have the app running anywhere publicly, just on my local machine. it's nowhere near production-ready and costs too much to query for me to expose it. this thing uses a lot more api tokens than i expected.

I used https://github.com/logan-markewich/llama_index_starter_pack/ to get started.

The notion code was actually in the llamaindex examples folder (i found this only after i'd cobbled something together off the internet).

https://github.com/jerryjliu/llama_index/blob/main/examples/data_connectors/NotionDemo.ipynb.

I have almost no experience with python, but managed to push through. definitely thinking about leveling up my python skills now. :)