Tavus

The Tavus Conversational Video Interface (CVI) meaningfully drives conversations forward with goal-oriented conversation management.

With Objectives & Guardrails, CVI can:

Define clear objectives with measurable completion criteria
Automatically branch based on user responses
Track progress, dependencies, and information in real time

This makes complex workflows, from health intakes to HR interviews to multi-step questionnaires, possible in conversational video experiences.

You can define Objectives & Guardrails in a few ways:

Automatically via the new Persona Builder
By creating your own JSON structure via the Objectives API

Objectives & Guardrails examples

An AI healthcare assistant can guide patients through a multi-step intake, adapting questions based on symptoms and responses.
An AI recruiter can conduct first-round interviews, staying on topic with role-specific questions and branching based on responses.
An AI trainer can conduct fitness assessments, where branching and logic tailors questions around experience level and fitness goals.

Ready to get started?

Guide CVI with clear goals and guardrails so it stays focused and effective, every time.

Get started

Start building:

Starter kit: Create an AI tutor

Learn how to create an AI tutor with Memories, RAG Knowledge Base, Guardrails, and more.

Starter kit: Create an AI sales assistant

Learn how to create an AI sales assistant with Memories, RAG Knowledge Base, Guardrails, and more.

Starter kit: Create an AI recruiter

Learn how to create an AI recruiter and interviewer with Memories, RAG Knowledge Base, Guardrails, and more.

FAQs

How do Objectives and Guardrails work?

With Objectives & Guardrails, Replicas can define objectives with measurable completion criteria, branch automatically based on user responses, and track progress in real time. This unlocks complex workflows like Health Intakes, HR Interviews, and multi-step questionnaires.