Voice AI platform that orchestrates the full voice agent stack — LLM, voice, transcriber, system prompt, structured outputs, tool registry, phone numbers and call analytics. Central building block for both inbound receptionists and outbound callers across Nate Herk’s voice agent tutorials.
Key facts
- Holds the system prompt, first message, model (GPT-4.1 or GPT-4o in the examples), voice and transcriber configuration
- Supports assistant overrides — runtime variable substitution inside the system prompt via {{variable}} syntax, so each call can receive dynamic context (lead name, company, request)
- Exposes structured outputs as a replacement for deprecated summary and success-evaluation fields; each field is typed (string, boolean, number) and can be reused across multiple assistants
- Provides up to 10 free US phone numbers per account; Twilio numbers can be imported to bypass outbound call limits
- Call lifecycle for outbound is two-step: POST to create a call, then GET on the call ID until status equals ended
- Supports a Handoff tool (assistant-to-assistant) and a Transfer Call tool (to external human phone numbers)
- Sends end-of-call reports to a configurable server URL with summary and structured data fields
- Pairs natively with n8n via either HTTP webhooks or an MCP server trigger
Source references
- [src-007] Nate Herk cluster — Nate Herk — Voice AI agents cluster (4 videos)
– Videos referenced: BO-jFbN4p8Y, y-cq_Qo4zVo, zWLZ3bVVwD8, Qt3zMBH-FNg
Recommended next
Keep reading from this thread
From 494 indexed pages and articles.
- Wiki concept Voice Agents AI agents whose primary interface is spoken conversation over a phone call, app, or in-browser audio stream. Related by vapi
- Wiki concept trigger.dev Background job platform used with Claude Code to host agents and workflows in production. Provides scheduled runs, automatic retries, queuing, orchestration, and a clean UI Readers have engaged with this next
- Insight AI Measurement and Experimentation How to measure AI product impact with evals, adoption metrics, online experiments, guardrails, and cost tracking Readers have engaged with this next