User Embeddings
The central concept of User Embeddings paves the way for the internet to evolve towards safer, more personalized experiences, overcoming critical obstacles. In an interconnected network of apps utilizing LLMs, user embeddings serve as a personal memory and knowledge repository that preserves privacy.
User Embeddings allow you to represent users as a set of points in embedding space. FirstBatch SDK provides the interface to seamlessly build and manage User Embeddings. The only prerequisite for applications is a vector database integration, meaning that items, content, or the search bar have underlying embeddings that can be retrieved fast.
How are User Embeddings created?
As users interact with the application, various types of signals with different weights and vectors are sent, collectively shaping User Embeddings. These embeddings are then converted to KNN queries by selected algorithms, which can be individually customized for each session and are fully managed by FirstBatch.
Users represented as vectors, what's the catch?
Representing users as a vector (or a set of vectors) gives remarkable flexibility for applications. Once a user has their own embeddings, FirstBatch can provide:
Navigable UX: "Search without a search bar"
Transform user experience into a navigable journey by leveraging user interactions. Every user action contributes to shaping their unique experience.
Individually Crafted Recommendations: "You may like this."
Offer users not only similar but also adjacent items in a personalized manner. This approach allows users to discover new and relevant content on their own terms, enhancing their exploration and satisfaction.
User-Intent AI Agents
Empower your AI agents with real-time insights into user intentions, derived from their interactions. This infusion of user intent brings intimacy to AI-driven experiences, making users feel more connected and understood.
Unique Journeys
Enable users to access the right content from the very beginning by tailoring their experience based on their starting point. This relies on something called a "persistent" session. Based on a user's previous embeddings, create journeys tailored for them.
User -Targeting Centric Promoted Content
Shift away from conventional targeting techniques and embrace a user-centric approach to deliver promoted items or ads in a captivating format. This approach allows users to actively influence the curation of promoted content, ensuring it aligns seamlessly with their preferences and resulting in a highly interactive and enjoyable experience.
Last updated
