Code: https://github.com/wgryc/phasellm/tree/main/demos-and-produc...
We're taking a proactive approach to our "chat with your database" product. We begin by inspecting your db, building a knowledge base, then AI will recommend relevant insights to you depending on your query history... tinder for data if you will!
To follow up just create a new request and it'll kickoff a chat
Basically, instead of text-to-SQL you do text-to-function-arguments, where that function does the analysis and the LLM translates natural language to your DSL instead of SQL or raw DB queries.
Shoot me an email if you want to explore this further (website in profile).
Check out this keynote from their last conference https://www.youtube.com/watch?v=9cMB7Sjzgt0