Cloud deployment

From your first call list to production

AI Rudder’s agents go live in six stages: designed with your team, connected to your systems, tested on real calls and taken live in stages, with a sign-off at every step.

Two ways to run AI Rudder

In our cloud, or in your data centre

The stages are the same. Where the agents run, who runs the infrastructure and how releases reach you are different.

Cloud

AI Rudder cloud

AWS or Alibaba Cloud, in the region you select
Run and monitored by AI Rudder
Scheduled releases, rolled out region by region
Agreed with you as your call volume grows
Suits
  • Lenders
  • Fintechs
  • BPOs
  • Service teams
On-premise

Your data centre

Your data centre or your own cloud account, for you alone
Your servers and GPUs, sized with AI Rudder; set up and kept current by AI Rudder’s engineers with your IT team
In your change windows, through your approval process
Set by the hardware you run
Suits
  • Banks
  • Regulated companies
How on-premise deployment works
Cloud deployment

From kick-off to go-live

Six stages, each with a sign-off by your team.

  1. 01Kick-offContract, project owners on both sides, the first set of calls
  2. 02Requirements and designCall flows, tone and scripts, written down and signed off
  3. 03Set-up and integrationAgents built in BotLab, connected to your systems and lines
  4. 04TestingInternal tests, then SIT and UAT with your team
  5. 05Go liveIn stages, with every call recorded and checked
  6. 06AcceptanceProject sign-off, then the service review

Then back to requirements for the next set of calls

What sets the timeline

At kick-off we agree a dated plan with you, and both sides track it every week. How long it takes depends on:

Call flows and languages

How many flows, in how many languages

The systems we connect to

Your CRM, core or loan system, and telephony

Third-party readiness

If a CRM, core system or carrier isn’t ready, the date moves even when our build is done

Security review and UAT

Your team’s security review and acceptance tests

Compliance sign-off

Per market

New, or one more scenario

A first deployment takes longer than one more set of calls on an account that is already live

Stage by stage

What happens, and who does what

01
Stage 01Kick-offEnds with a dated plan

Your team

  • A project owner and a business owner
  • The goals and limits for the first calls

AI Rudder

  • A project lead and in-market engineers
  • A dated project plan that both sides track every week
What happens
  • Contract signed, project started
  • Owners named on both sides, and one place to raise requests
  • The first set of calls chosen: which calls, which market, which rules
02
Stage 02Requirements and designEnds with your sign-off

Your team

  • Your rules: calling hours, consent, disclosure, handoff
  • Sample data and the fields each system holds
  • Sign-off on the design

AI Rudder

  • The solution design
  • The call flows mapped to your rules
What happens
  • A business analysis workshop on your calls
  • Call flows, tone of voice and scripts, in your customers’ language
  • A written specification for every integration, with sample files (file formats, paths, callback addresses)
03
Stage 03Set-up and integrationPaced by the systems we connect

Your team

  • Access to your systems’ APIs, or an SFTP server for file exchange
  • Your carrier lines or SIP trunk, and the numbers to call from
  • IP addresses to allow, and single sign-on if you use it
  • Your vendors ready on time: CRM, core system and carrier

AI Rudder

  • Agent configuration and voices
  • The integrations, built to the signed specification
  • WhatsApp Business messaging through an official provider, where you need it
What happens
  • A test environment in AI Rudder’s cloud, in your region
  • Agents built and trained in BotLab
  • Connections to your CRM, core or loan system and telephony, by API or by file transfer
04
Stage 04TestingPaced by your UAT and security review

Your team

  • Test data and test cases
  • Your sign-off on UAT

AI Rudder

  • Test runs, and fixes turned around inside the UAT window
  • The move to production once UAT is signed off
What happens
  • Internal tests of every flow and integration
  • System integration testing (SIT) on your test data
  • User acceptance testing (UAT): real test calls your team listens to and approves, with a parallel run if you want one
05
Stage 05Go liveThen in stages

Your team

  • Decide when to add volume, portfolios or markets

AI Rudder

  • Monitoring, alerts and daily checks while volume grows
What happens
  • Production starts with one set of calls, then more volume
  • The first days after each step are watched closely
  • Every call recorded, transcribed and checked against your rules
06
Stage 06Acceptance and serviceAfter go-live

Your team

  • Acceptance of the project
  • Changes to scripts and rules in BotLab, without us

AI Rudder

  • A named customer success manager
  • Support to the response times below
What happens
  • Project sign-off
  • Service reviews with your customer success manager
  • New use cases and markets planned from what the calls show
After go-live

How changes reach you on cloud

AI Rudder ships updates on a release calendar, so your agents change on a known schedule.

A release calendar

Scheduled releases, rolled out one region at a time

Watched after every release

Close monitoring the day after each update reaches your region

Urgent fixes

Outside the calendar, when they can’t wait

Your requests go to one queue. Each is written up, estimated and scheduled into a release, and you approve it in UAT before it reaches production.

Support

Response times on cloud

SeverityFirst responseQuick fixPermanent fix
Critical1 hour5 hours5 business days
High2 hours8 hours10 business days
Medium4 hours24 hours25 business days
LowAgreed with youAgreed with youNext maintenance release

Quick fixes are measured around the clock; permanent fixes in business days. These times are a guide: the response times in your contract, agreed with you, are the ones we commit to.

What you control

What your team changes without us

Your team sets the rules, the scripts and the limits, and changes them in BotLab.

Scripts and call flows

In BotLab

Rules and limits

Per deployment

Consent and disclosure

Per market

Handoff to people

When, and to whom

Knowledge sources

Knowledge Studio

Responsible AI

Responsible AI on every call

How Responsible AI works
Your rulesScripts, limits and contact times set by your team
Disclosure and consentPer market, on every call
Handoff to peopleWith the context, when a person is needed
A full recordRecording, transcript and QA check for each call

Need the agents inside your own data centre?

On-premise deployment

Tell us which market, and which calls.

Our in-market team will plan the first deployment with you, in the language your customers speak.