Fine-grained segmentation of content items for tiered cloud archival using machine-learned models

A computing system comprising:

• a memory;
• one or more processor devices configured to:
• responsive to a request to store a content item for a user of a content service, process user metadata and content item metadata with a machine-learned embedding model to obtain a query embedding;
• perform a vector similarity search on a vector database with the query embedding to retrieve a stored embedding matching the query embedding, wherein the stored embedding represents a contextual data element related to playback of the content item;
• process a set of inputs comprising the contextual data element, the user metadata, and the content item metadata with a machine-learned model to obtain an item storage decision output that identifies a segmentation point at which to segment the content item into a first segment and a second segment, and wherein the item storage decision output further identifies a first storage repository for the first segment and a second storage repository for the second segment;
• and
• store the first segment to the first storage repository and the second segment to the second storage repository.

A non-transitory computer-readable storage medium that includes executable instructions to cause one or more processor devices to:

• responsive to a request to store a content item for a user of a content service, process user metadata and content item metadata with a machine-learned embedding model to obtain a query embedding;
• perform a vector similarity search on a vector database with the query embedding to retrieve a stored embedding matching the query embedding, wherein the stored embedding represents a contextual data element related to playback of the content item;
• process a set of inputs comprising the contextual data element, the user metadata, and the content item metadata with a machine-learned model to obtain an item storage decision output that identifies a segmentation point at which to segment the content item into a first segment and a second segment, and wherein the item storage decision output further identifies a first storage repository for the first segment and a second storage repository for the second segment;
• and
• store the first segment to the first storage repository and the second segment to the second storage repository.