veevo.ai
Deploy AI phone agents in minutes, not months. Bring your own Twilio number and OpenAI key.We handle the rest.
How It Works
Features
Twilio and OpenAI credentials are returned per-call via your onCallStart callback. Veevo never persists them.
Define OpenAI function-calling tools. Veevo routes invocations to your backend mid-conversation via onToolCall.
Gate callers with an allowlist. Rejected calls cost ~$0.01 vs ~$0.10 for a full AI session.
onCallEnd delivers the complete transcript plus a token-level cost breakdown for every call.
Initiate AI-powered outbound calls via a single REST endpoint. Same callbacks, same control.
Every callback carries an X-Veevo-Signature header so you can verify requests before returning credentials.
Integration
onCallStart — return your credentials + config when a call arrives.
onToolCall — handle function calls from the AI mid-conversation.
onCallEnd — receive the full transcript and cost breakdown.
app.post("/on-call-start", (req, res) => {
const { callerNumber } = req.body;
// Reject unknown callers
if (!allowlist.includes(callerNumber)) {
return res.json({
reject: { message: "This number is not accepting calls." }
});
}
res.json({
twilioAccountSid: process.env.TWILIO_SID,
twilioAuthToken: process.env.TWILIO_TOKEN,
openaiApiKey: process.env.OPENAI_KEY,
systemPrompt: "You are a helpful assistant.",
onToolCallUrl: "https://api.example.com/on-tool-call",
voice: "shimmer",
tools: [{
name: "check_order",
description: "Look up an order by ID",
parameters: {
type: "object",
properties: {
orderId: { type: "string" }
},
required: ["orderId"]
}
}]
});
});Pricing
$0/mo
+ $0.00/min
You pay OpenAI and Twilio directly. No markup on third-party costs.
FAQ