QM by AI Deployment Guide

Introduction

This guide explains the Kubernetes deployment of Quality Management by AI in EFCX. Before you begin, ensure EFCX is fully installed and you have root access to your Kubernetes cluster and Helm. Additionally, you must have Vault access to configure Redis credentials and the necessary API keys/endpoints for your chosen LLM provider (e.g., OpenAI, Gemini, Azure).

Ensure you are using CX version 5.5.0 or above and are currently inside the Active CX /kubernetes directory before proceeding with the deployment.

1. Set up Redis

Kindly follow the Agent Co-Pilot Deployment Guide | 1. Set up Redis guide to install Redis and configure Vault. If you’ve already installed Agent Co-Pilot, skip this step.

2. Configure Vault Access for Redis Credentials

Kindly follow the Agent Co-Pilot Deployment Guide | 2. Configure Vault Access for Redis and AMQ Credentials guide to configure Vault. If you’ve already installed Agent Co-Pilot, skip this step.

3. Apply Secrets and ConfigMaps

3.1. Apply secrets in your tenant namespace

Apply the SecretProviderClass:

kubectl -n <YOUR_TENANT_NAMESPACE> apply -f pre-deployment/expertflow-ai/expertflow-ai-vault.yaml

Verify:

kubectl -n <YOUR_TENANT_NAMESPACE> get secretproviderclass expertflow-ai-vault

Copy Redis TLS secret to YOUR_TENANT_NAMESPACE:

kubectl get secret redis-thirdparty-crt -n ef-external -o yaml \
  | sed "s/namespace: ef-external/namespace: <YOUR_TENANT_NAMESPACE>/" \
  | kubectl create -f -

Copy Image pull secret to YOUR_TENANT_NAMESPACE:

kubectl get secret expertflow-reg-cred -n ef-external -o yaml | sed "s/namespace: ef-external/namespace: <YOUR_TENANT_NAMESPACE>/" | kubectl create -f -

3.2. Apply secrets in the expertflow namespace

Kindly follow the Agent Co-Pilot Deployment Guide | 3. Apply Secrets and ConfigMaps guide to apply secrets and ConfigMaps for expertflow namespace. If you’ve already installed Agent Co-Pilot, skip this 3.2 step.

4. Deploy QM by AI

4.1. Install Agent Co-Pilot

If you’ve already installed Agent Co-Pilot, skip this 4.1 step.

You must first create a custom values file to define your minimum required configurations.

Run the following command to create a new file:

vi helm-values/cx-agent-co-pilot-custom-values.yaml

In the opened file, add the following section to define the FQDN (Fully Qualified Domain Name) for ingress routing:

global:  
  ACTIVEMQ_PRIMARY_URL: <IP-ADDRESS>
  ACTIVEMQ_SECONDARY_URL: <IP-ADDRESS>

llm-connector:
  enabled: true

ai-analyzer:
  enabled: false

🔁 Replace <IP-ADDRESS> with your actual ACTIVEMQ IP address.

Now follow the Agent Co-Pilot Deployment Guide | 4. Deploy Agent Co Pilot guide to deploy Agent Co-Pilot using this helm-values/cx-agent-co-pilot-custom-values.yaml file.

4.2. Install QM by AI

You must first customise the values file to define your minimum required configurations.

Run the following command to open the file:

vi helm-values/mtt-single-custom-values.yaml

In the opened file, add the following section to define the FQDN (Fully Qualified Domain Name) for ingress routing:

global:
  tenantId: <tenant-id>
  ingressRouter: <CUSTOM-FQDN>

ai-evaluator-fetcher:
  enabled: true

ai-evaluator-processor:
  enabled: true

🔁 Replace <CUSTOM-FQDN> with your actual domain, e.g., ai.example.com. Replace <tenant-id> with your actual tenant ID.

Finally, deploy the QM by AI using the following command.

helm upgrade --install --namespace <YOUR_TENANT_NAMESPACE> --debug  <tenant-id> --values helm-values/mtt-single-custom-values.yaml  expertflow/MTT-single --version 5.5.0