The scenario loads documents from Google Drive into a Postgres PGVector Store base and answers questions strictly from them, checking MISP for bookings.
Answers from your documents, no making up
Search by meaning, not exact words
Accounts for calendar availability
The customer gets an accurate answer
MISPPostgres PGVector StoreGoogle DriveWebhookQdrant Vector StoreEmbeddings OpenAIDefault Data Loader