Coru
Book a demo

ADI AGENTS · TRANSACTIONS

Voice-authenticated operations, in seconds

Transfers, enquiries and payments executed inside the app, with authentication and a full record of every step.

A closed action catalogue, with enums validated outside the model
The agent never receives the end customer's credentials — it calls your authorised endpoint
Responses under 800 ms between instruction and confirmation
Every operation produces a verifiable event, signed with HMAC-SHA256
ISO 27001 PCI SOC 2
14:32 ▮▮▮ ⌁ 84%
ADI · Example Bank online
Hello, Marcos. I am ADI. How can I help today? 14:30
i want to pay the electricity bill 14:33
Message

The operational jump

From information to execution

Both models handle the same request. What changes is who completes the operation.

An assistant that only informs

× It looks things up and explains; when the customer wants to execute, it hands over a link
× The experience gain is lost at the moment of highest value
× The traditional channel keeps carrying the same volume
× A tree-shaped IVR makes the customer guess where their intent lives
× Latency above 800 ms makes the customer repeat themselves

ADI transactions agent

Executes within a closed catalogue, with enums validated outside the model
HTTP Tools pointing at your APIs, under your authorisation policy
Responses under 800 ms on our own voice stack
Amounts and receipts pronounced from a dedicated dictionary
A verifiable event by signed webhook on every operation

Results in production

Impact that is measured

Measured on our own voice stack, with no chained third-party APIs.

<800 ms time to first byte of a voice response
70% of volume automated, with human relief on demand
1–3 weeks from defined case to production

How it runs

From instruction to receipt

01 · SESSION

Authenticated customer

The conversation starts inside the app or on an authenticated channel, with the session already validated.

02 · INTENT

Request in free speech

The customer says what they want to do, with no menu and no numbered option to pick.

03 · CATALOGUE

Permitted action

The operation resolves to a registered enum. What is not on the list does not happen.

04 · AUTHORISATION

Confirmation and policy

The agent confirms details and calls your endpoint, which operates under your authorisation policy.

05 · EXECUTION

Operation completed

Transfer, enquiry or scheduling executed through HTTP Tools against your APIs.

06 · RECORD

Verifiable event

A receipt to the customer and an event signed with HMAC-SHA256 to your system.

Capabilities

Executing without opening a gap

Closed action catalogue

Transfer, check balance, schedule a payment. What is not registered does not happen, whatever the request.

No credentials in the agent

The agent never receives the end customer's credentials; it calls an endpoint authenticated by API key or header.

Correctly pronounced amounts

The agent reads "one thousand two hundred thirty-four reais and fifty centavos", not the raw string.

Encrypted call end to end

SIP TLS 5061 with SRTP, including the telephony layer.

VAD barge-in

The customer interrupts and the agent yields the turn, as in a real conversation.

An event per operation

Every executed action produces a signed webhook, traceable in your system.

Feasibility

Connects to what you already have

Core, transactional APIs and telephony connected through a documented interface. No core changes.

In-app voice widget PSTN voice WhatsApp Business REST APIs · JWT HMAC webhooks HTTP Tools MCP SIP TLS · SRTP
Timeline

Onboarding in 1 to 3 weeks. With an existing SIP trunk, the voice channel goes live in 2 to 5 days.

Inside the rules

Transacting without leaving policy

Authorisation on your side

The endpoint called already operates under your policy; the agent does not decide permissions.

Guardrails in code

Deterministic filters before and after the model, three of them mandatory on any agent.

Per-turn traceability

What the agent said, on which version, under which controls, and which action it executed.

Tokenised PII

Personal data is tokenised before any call to a model provider.

FAQ

Frequently asked questions

Can the agent execute an operation we did not authorise?

No. Every available action is registered as an enum validated outside the model. What is not on the list does not happen, whatever the customer asks or however they ask it.

Does the agent have access to customer credentials?

No. It calls an endpoint of yours, authenticated by API key or header, which already operates under your authorisation policy.

How do we audit each operation?

Every operation produces a verifiable event, delivered by a webhook signed with HMAC-SHA256, and the interaction stays reconstructable with the agent version and the active controls.

Is voice secure enough to transact?

The call travels encrypted, including the telephony layer, over SIP TLS 5061 with SRTP. Session authentication happens before any amount is mentioned.

Why does 800 ms matter?

Above that threshold the customer notices the wait between instruction and confirmation and starts repeating themselves. The voice stack is our own precisely so latency is controlled in one place.

Request a demo

Bring your authorisation policy.

The agent's action catalogue is built from it, before the pilot.

What you bring

Your authorisation policy, the operations you want covered and the current volume on the traditional channel.

What we bring

The action catalogue configured with validated enums, our own voice stack and a dedicated compliance liaison.

What comes out

A transactional agent in sandbox, a proposed pilot KPI and the basis of the rollout plan.

Shall we start?

Please use your work email.

By submitting you accept our privacy policy. We do not share your information with third parties.

Newsletter

Insights on AI in regulated operations, once a month.

Governance, architecture and what happens after go-live. No spam.

By subscribing you agree to receive communications from Coru.