LiveKitReal‑time presence
Powers Shaggy’s live audio and video presence so conversations feel like a real-time call, not a loading bar.
The technologies that make a digital human feel human
Shaggy. To be able to replicate the human experience is a goal born out of science fiction. Shaggy is a live running benchmark built collaboratively by leading AI labs and startups to see if they can bring this vision to life.
Who's powering Shaggy right now?
Each company below owns a slice of Shaggy's brain, body, or memory. Together they form the stack that lets him see, speak, remember, and improvise like a person.
SimultaneousMulti‑agent coordination
Lets multiple models and tools work in parallel behind the scenes while Shaggy handles just one friendly conversation.
Mem0Long‑term memory
Gives Shaggy a memory you can grow over time—so he can remember people, projects, and promises, not just single prompts.
AgentMailOutreach & follow‑ups
When Shaggy says he’ll follow up, AgentMail is how the message actually lands in your inbox at the right moment.
AnamFace & embodiment
Renders Shaggy as a living avatar on screen, giving his words a face, body, and physical presence instead of just chat bubbles.
CartesiaVoice
Gives Shaggy a voice that can shift tone, pacing, and emotion so he sounds like a person, not a system prompt.
DeepgramListening
Handles speech recognition so Shaggy can listen in messy, real‑world audio and still respond like he caught every word.
EncountrEmotional analysis
Reads emotional tone from Shaggy’s voice interactions so he can respond not just to what people say, but how they feel when they say it.
SinchChannels
Connects Shaggy to the outside world—phones, messaging, and communication rails—so he can reach people where they actually are.