Voice AI
Agents that talk with people on your website, in your app, or over a phone
call. Speech in, speech out, fast enough to hold a conversation.Start here
Physical AI
Robots that a person or a model drives over the internet. Actions and
camera frames, checked against your safety limits and recorded.Start here
Why Zero Runtime
Real code, not a dashboard
Ordinary Python, running in your process with your own libraries and data.
Your providers, your hardware
Swap any speech-to-text, LLM or voice in a line. Connect any robot.
A worker, not a cluster
You deploy one worker. Nothing else is yours to run.
Voice AI use cases
Customer support
A support agent that hands off to a person over MCP instead of guessing.
Appointment scheduling
A phone agent that books slots. Its tools are an n8n workflow reached over
MCP.
Live translation
Detects the caller’s language on every turn and rebuilds the pipeline to
answer in it.
Multilingual support
Detects the user’s language and answers in the one they prefer.
Physical AI use cases
Remote teleoperation
Drive a robot from another room or another country, inside the safety
limits you set.
Data collection
Record a person driving, then export a LeRobot dataset. Each action is
paired with the camera frame that caused it.
Simulation and policy
Neither side has to be real hardware. Drive a simulated robot, or let a
model send the movements.
Supervised autonomy
Let a model drive, and have a person take over whenever they need to.