Connect your voice agent with Notion to automatically store structured information captured during phone conversations. The integration enables agents to create pages, update records, and save call summaries directly in your Notion workspace.
This video shows how a Voice AI agent built on Telnyx voice captures information during a conversation and writes structured data into a Notion database. The agent records key details, creates a new page, and tags it - all without manual data entry.
The Greenhouse integration supports the following operations across core objects:
Candidates
Applications
Jobs
Users
Notion is an AI-powered workspace that brings notes, documents, databases, and project management into a single system of record. Teams use it to capture knowledge, find answers, and automate tasks with AI agents built for collaborative work.
A Telnyx voice AI agent can search, create, and update content in Notion through API tool calls during a phone conversation. When a caller asks a question, the agent can search Notion pages and databases for relevant information and read back the results in natural language.
The agent can create new pages or database entries from call context - logging meeting notes, capturing action items, or recording customer feedback into a Notion database. It can update existing pages by appending content, changing property values, or adding the call summary to a project tracker.
For knowledge retrieval workflows, the agent can query Notion databases by property filters, retrieve page content, and synthesize answers from multiple sources - turning a Notion workspace into a voice-accessible knowledge base that callers can interact with hands-free.
COMPOSITION
Four Notion assistants. Pick one and the blocks it runs on light up, and the create call attaches Notion in the same request, the meter prices it. Or click any single block: every primitive sells alone.
One request. Tools attached. You pay the minute, not the connector.
Every Telnyx integration runs on the same infrastructure: voice AI agents for conversation, inference for reasoning, SIP trunking for transport, STT and TTS for understanding and speech, and phone numbers for provisioning.

The conversational layer that connects to Greenhouse, handles the call, and triggers tool calls to create candidates and update applications.
Learn more
Open-weight LLMs power the reasoning behind every tool call, so the agent knows when to create a candidate versus update an existing application.
Learn more
HD audio over SIP delivers clear call quality so the agent captures every detail a candidate shares, even on a noisy phone line.
Learn more
Real-time transcription converts what candidates say into text the agent can reason over and act on inside Greenhouse.
Learn more
Natural voice synthesis lets the agent ask screening questions and confirm candidate details in a conversational tone.
Learn more
Provision a dedicated recruiting line in seconds. Every Greenhouse integration needs a number for candidates to call.
Learn moreOptimize recruiting with voice-driven updates, scheduling, and candidate workflows.
Learn moreCHOOSE BY USE CASE OR PRODUCT
RESULTS
cURL
# voice-powered knowledge base — create the agent# tools attach in the same requestcurl -X POST https://api.telnyx.com/v2/ai/assistants \ -H "Authorization: Bearer $TELNYX_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "name": "notion-knowledge-base", "model": "meta-llama/Meta-Llama-3.1-8B-Instruct", "instructions": "Answer the call, ask the caller what they need to know, search Notion for relevant pages, and read out the answer.", "integrations": [ { "integration_id": "notion" } ] }'# 7 primitives · 1 vendor · 1 bill# 1 tool · GET /v2/ai/integrationsVoice-Powered Knowledge Base
SELECT USE CASE
SELECT USE CASE
OR SELECT PRODUCT
RESULTS
cURL
# voice-powered knowledge base — create the agent# tools attach in the same requestcurl -X POST https://api.telnyx.com/v2/ai/assistants \ -H "Authorization: Bearer $TELNYX_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "name": "notion-knowledge-base", "model": "meta-llama/Meta-Llama-3.1-8B-Instruct", "instructions": "Answer the call, ask the caller what they need to know, search Notion for relevant pages, and read out the answer.", "integrations": [ { "integration_id": "notion" } ] }'# 7 primitives · 1 vendor · 1 bill# 1 tool · GET /v2/ai/integrationsVoice-Powered Knowledge Base
RESULTS
cURL
# voice-powered knowledge base — create the agent# tools attach in the same requestcurl -X POST https://api.telnyx.com/v2/ai/assistants \ -H "Authorization: Bearer $TELNYX_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "name": "notion-knowledge-base", "model": "meta-llama/Meta-Llama-3.1-8B-Instruct", "instructions": "Answer the call, ask the caller what they need to know, search Notion for relevant pages, and read out the answer.", "integrations": [ { "integration_id": "notion" } ] }'# 7 primitives · 1 vendor · 1 bill# 1 tool · GET /v2/ai/integrationsVoice-Powered Knowledge Base