Skip to content →
    • Log in
    • Start Free

Australia

Voice agents, built and run in Australia.

The number is Australian. The model, the speech and the recordings run in Sydney. You write what the agent says and knows, and Meddle runs everything under it.

Start buildingAnswering inbound calls

Where it runs

Onshore, in four numbers

Data at rest

Sydney

AWS ap-southeast-2

Model inference

Sydney

AWS Bedrock

Voice pipeline

Sydney

Fly.io

Built in

Perth

Vertical AI Pty Ltd

What you are choosing

A platform, not a phone-answering service

You own it

The agent is yours to change

Nobody rewrites your prompt on your behalf, and there is no agency between you and the call. Open the editor, change a line, publish. The version that answers is the version you published.

Swap any part

Model, listener, voice

Change the language model, the speech recognition or the voice per agent, and calls keep running. You are not locked to one vendor's stack because a salesperson bundled it.

Read the calls

Every call, turn by turn

Transcripts, timings and scores for real calls, not a monthly summary. Most fixes are one line in the prompt, and reading the call is how you find which line.

Prove it first

Scored before it launches

Scenarios you write, graded by a judge, run on demand. The failing call shows up in your test results instead of in a complaint.

Hand it over

A person is one tool call away

Straight hand-off or a screened warm transfer that announces the caller. Configured per agent, and honest with the caller when no destination is set.

Describe, do not configure

Build it by talking to it

Everything in the editor is reachable from the AI assistant you already use, so you can set an agent up and change it in plain language instead of learning a console.

How you build one

Five steps, in the order you would guess

  1. 01

    Describe the job

    Write what the agent says, what it knows, and the rules it follows. Plain sentences, not a flowchart. Do it in the editor, or by describing it to the AI assistant you already use.

  2. 02

    Give it something to look up

    Point it at your prices, your hours, your policies. It reads from that instead of inventing an answer, and says so when the answer is not there.

  3. 03

    Give it something to do

    Book the job, send the text, check the order, pass the call to a person. Each one is a tool the agent can reach for mid-call, and you choose which.

  4. 04

    Test it until it stops surprising you

    Write the calls you dread as scenarios and play them at the agent. A judge grades each run, so you find the one that fails before a caller does.

  5. 05

    Put it on a number

    Buy an Australian inbound number in the workspace and route it to the published version, or keep the number you have and forward it. Every call after that is scored and readable turn by turn.

Questions

What Australian buyers ask first

Where does an Australian call actually get processed?
Call data at rest, meaning transcripts, agent configuration, embeddings and audit logs, sits in Supabase Postgres on AWS ap-southeast-2 in Sydney. Model inference runs on AWS Bedrock in Sydney. The voice pipeline runs on Fly.io in Sydney. Speech recognition and synthesis are reached over TLS from providers whose agreements forbid training on your input; the full provider-by-provider breakdown, including the paths that leave Australia, is published at docs.meddle.sh/docs/trust/data-residency.
Do I need a new phone number?
No. You can buy an Australian inbound number inside the workspace and route it to your agent, or keep the number your customers already have and forward it: all calls, after a set number of rings, or only when you are already on a call. Porting the number across is also supported if you would rather move it.
Can the agent put someone through to a human?
Yes, and it is a first-class tool rather than a fallback. With a transfer destination configured the agent dials it, either as a straight hand-off or as a screened warm transfer that announces the caller first. With no destination configured it says plainly that it cannot connect them and keeps helping rather than pretending.
How do I know it works before I point real callers at it?
You write the calls you are worried about as scenarios and run them. Each run is graded by a language-model judge against the rubric you set, so a change is proved before it ships rather than after a customer meets it. Every real call is scored on the same metrics once it is live.
Who am I buying from?
Vertical AI Pty Ltd, ABN 73 695 607 946, based in Perth. Meddle is the product. Support and legal paperwork, including a DPA, come from the same entity.
Does this work for outbound calling too?
Outbound campaigns exist, and launching one requires you to confirm the list complies with ACMA rules and the Do Not Call Register. The platform will not dial a list without that confirmation. Inbound is the surface most operators start on.

Build one and call it.

The fastest way to judge a voice agent is to ring it. Build yours, place a test call, and read the transcript.

Start building

Platform

  • Voice agents in Australia
  • Inbound calls

Compare

  • Meddle vs Retell
  • Meddle vs Bland
  • Meddle vs Synthflow

Docs

  • Getting started
  • Build an agent
  • Testing agents
  • Data residency

Legal

  • Security
  • Privacy
  • Terms
  • Sub-processors
  • DPA
© 2026 MeddleReading this as a model? Start at /llms.txt, or ask /llms.