Chert provides AI-powered video agents designed to conduct FaceTime calls, enabling automated support and interaction by processing visual context from the caller's camera.
What the product does and how it is positioned
Chert offers a platform for building AI agents that can see and interact through FaceTime. By leveraging camera context, these agents are designed to assist in scenarios where visual information is necessary for guidance.
The platform allows for the configuration of agent personas, including voice, avatar, and framing. Users can manage these agents through a control plane that supports the deployment of assistants to provisioned FaceTime lines.
Source-supported ways to use the product
Agents guide callers through technical issues by viewing the environment or equipment through the caller's camera.
The platform facilitates intake processes or visual assessments by allowing agents to observe and respond to visual inputs in real-time.
The documented workflow, where available
Define the agent's job, specific instructions, underlying model, voice, and behavior before initiating a call.
Choose the appropriate avatar and visual framing to match the desired user experience.
Publish the assistant and assign it to a provisioned FaceTime line to enable live interaction.
Chert distinguishes its agents from traditional voice-only AI by incorporating visual processing capabilities. This allows the agent to act as a participant in a FaceTime call, observing what the caller shows to provide context-aware guidance.
Checks to run with your own material and workflow
What was checked and when
Answers based on the source-checked product record
No, the FaceTime product is currently available through a controlled private preview while the company validates production media and line readiness.
Yes, the browser preview allows you to check the prompt, voice, microphone, interruption behavior, avatar, and cleanup processes before publishing.
The control plane supports bounded inbound and outbound test workflows. Live execution is provisioned and explicitly authorized; accepting mode is not enabled by default.
You can configure the assistant's instructions, the underlying realtime model, voice, avatar, framing, publishing state, and managed line assignment.
Yes, the company provides stable iMessage infrastructure for inbound and outbound messaging via API and webhooks.
