I don't have any experience trying to store ML data in JSONB. Currently I can think of two places we use them at Threads Styling:
We have a "MediaStore" service for storing and processing images and videos. This service is consumed by a range of different services internally, so we allow storing a small blob of metadata along with each image. This makes it really flexible for supporting lots of different scenarios.
We have a service that synchronises contacts between google contacts and an internal database (for our employees). We store the data we currently care about (name, phone numbers etc.) in normalised columns in postgres, but we also dump the entire JSON document for each contact into a JSONB column, which has proved useful for debugging, and would allow us to extract extra info if we needed by running a db migration, rather than needing to re-sync with google.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
I don't have any experience trying to store ML data in JSONB. Currently I can think of two places we use them at Threads Styling:
We have a "MediaStore" service for storing and processing images and videos. This service is consumed by a range of different services internally, so we allow storing a small blob of metadata along with each image. This makes it really flexible for supporting lots of different scenarios.
We have a service that synchronises contacts between google contacts and an internal database (for our employees). We store the data we currently care about (name, phone numbers etc.) in normalised columns in postgres, but we also dump the entire JSON document for each contact into a JSONB column, which has proved useful for debugging, and would allow us to extract extra info if we needed by running a db migration, rather than needing to re-sync with google.