Home Use cases Semantic search

Search that understands a customer’s own words

People type the words that happen to be in a record, and miss the ones that mean the same thing.

What the user sees.

A customer types a question the way they would ask a colleague. They see matches from their own account, not a keyword dump from the whole database.

How that customer’s data stays put.

We embed that tenant’s records and filter by tenant before anything is returned. Another customer’s rows never enter the result set.

The build is RAG for a multi-tenant product, and the boundary is the same one in the architecture note. It also sits under AI features for SaaS.

FAQs

Questions that come up

Yes. Any one of these can be the first feature, once the product and the tenant boundary are clear.

Related

logo