A few more additional thoughts while anyone is looking:
I hope to get a fake demo dataset made soon. It doesn't seem all that useful when you login with nothing there. Similarity search is pretty meaningless until you have things in there. And it's cool to get a sense of how notes can build over time to something useful. As a personal crm, I'm not sure what would be easier.
Also need to make a more detailed video describing use cases, and explaining what the basic data models are.
The last day or so I've been working on more structured data extraction -- as I have more fun doing that than trying to fix a few regex issues on the front end.
fyi, I deployed a bug last night that broke the calendar. Fixed and deployed right now.
Thanks anyone for looking!
Really inspiring. Let it bloom, and keep it up!
Gpt-3 for automatic entity extraction/tagging. Embeddings and Pinecone for similarity searches.
Suggestion:
Record a new demo video with what I am going to call "normal" contrast white background and light elements theme. Your demo is super low contrast and very dark. Hard to watch.
Take cues from the way UI colors work on computers as shipped to the average person. Nobody ships their OS with super dark low contrast everywhere. Just suggesting, if you intend to actually go to market, you might just turn off your average user.