The social intelligence layer for AI

Citrus is a model-driven interaction stack that adds social intelligence to frontier AI systems. Specialized perception, user-modeling, and steering components interpret people over time and adapt AI behavior in real time.

Interpret What is happening with this user right now?
→
Understand Who is this user, and what matters to them?
→
Steer How should the AI behave next?
Real-time perception

Interpretation Layer

Models the interaction as it unfolds across language, audio, vision, engagement, and software event context. This captures not only what a user says but also how they say it and which moments matter.

SignalsMultimodal behavior and interaction context.
Temporal ContextTracks how interaction state changes across moments and turns.
StateProduces structured context downstream systems can act on.
Persistent user modeling

Rapport Layer

Builds a persistent representation of the person behind the interaction, separating relatively stable preferences and baselines from transient state so context can evolve with the user over time.

PreferencesLearns how users prefer to communicate and respond.
BaselinesTracks individual norms so behavior is interpreted relative to the person.
ContinuityMaintains longitudinal context across interactions instead of starting from zero.
Adaptive behavior

Steering Layer

Uses live interaction state and persistent user context to adapt how downstream AI behaves toward application objectives and constraints. Modulated AI behavior can include its timing, communication style, strategy, and task execution.

StrategySelects how the AI should approach the interaction based on the user and objective.
ExecutionAdapts language, timing, behavior, and task flow in real time.
ConstraintsCan incorporate product, safety, and user-level boundaries into the steering objective.