# Installation DuploCloud AI Suite is installed directly into your cloud account, running inside a Kubernetes cluster that you own and control. This document explains what you need to provide to DuploCloud, what we will set up, and what prerequisites your environment needs to meet before installation begins. *** ## How Installation Works DuploCloud deploys the AI Suite into your cloud environment using a Helm chart. We manage the deployment process on your behalf — you do not need to run any commands yourself. Once the prerequisites are in place and access is granted, our team handles the installation end-to-end. *** ## Supported Cloud Providers DuploCloud AI Suite runs on any of the following managed Kubernetes platforms: | Cloud | Kubernetes Service | | ---------------- | ------------------ | | **AWS** | Amazon EKS | | **Google Cloud** | GKE | | **Azure** | AKS | *** ## What You Need to Provide ### 1. A Kubernetes Cluster You must have a managed Kubernetes cluster running in your cloud account. DuploCloud can assist with cluster setup if one does not already exist. * **AWS:** Amazon EKS cluster * **GCP:** GKE cluster * **Azure:** AKS cluster ### 2. Cloud Infrastructure Depending on your cloud provider, the following infrastructure must be provisioned in your account before installation: #### AWS (EKS) | Resource | Purpose | | -------------------------------- | ----------------------------------------------------------------------- | | **AWS Load Balancer Controller** | Routes external traffic to the application | | **EFS Filesystem** | Shared file storage for the backend (mount targets in each node subnet) | | **EFS CSI Driver** | Connects the cluster to EFS | | **ACM Certificate** | TLS for the application domain | | **S3 Bucket** | Object storage for AI-generated files and reports | | **IRSA Role** *(optional)* | IAM role for the backend service account with S3 + EFS permissions | #### Google Cloud (GKE) | Resource | Purpose | | ------------------------ | ------------------------------------------------- | | **Ingress Controller** | Routes external traffic (nginx or GKE built-in) | | **Filestore CSI Driver** | Shared file storage for the backend | | **GCS Bucket** | Object storage for AI-generated files and reports | | **Workload Identity** | Grants the backend service account access to GCS | #### Azure (AKS) | Resource | Purpose | | -------------------------------- | ------------------------------------------------------------ | | **Ingress Controller** | Routes external traffic (nginx or Azure Application Gateway) | | **Azure Files CSI Driver** | Shared file storage for the backend (built-in on AKS ≥ 1.21) | | **Azure Blob Storage Container** | Object storage for AI-generated files and reports | | **Azure Workload Identity** | Grants the backend service account access to Blob Storage | ### 3. A Domain Name and TLS Certificate You must provide: * A **public domain name** (or subdomain) where the application will be hosted, e.g. `ai-helpdesk.yourcompany.com` * A valid **TLS certificate** for that domain (ACM on AWS, or cert-manager/Let's Encrypt on GCP/Azure) ### 4. Google OAuth Credentials DuploCloud AI Suite uses Google OAuth for user authentication. You must provide: * A **Google OAuth Client ID** * A **Google OAuth Client Secret** These are created in the [Google Cloud Console](https://console.cloud.google.com/) under **APIs & Services → Credentials**. The authorized redirect URI should point to your application domain. ### 5. Administrator Email Addresses Provide a list of email addresses that should have **super-admin access** to the platform. These users will have full administrative control and can manage other users after installation. *** ## What DuploCloud Installs Once the prerequisites are in place, DuploCloud deploys the following into your cluster: | Component | Description | | ----------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- | | **Backend** | ASP.NET Core API server that powers the AI HelpDesk, project management, and agent orchestration | | **Frontend** | React web application served via nginx — this is the UI your team interacts with | | **MongoDB** | Embedded database (Kubernetes StatefulSet) for storing tickets, projects, reports, and configuration | | **Web Terminal** *(optional)* | Browser-based terminal integration for agent command execution | | **Ingress** | Load balancer / ingress resource routing HTTPS traffic to the application | | **Persistent Storage** | Shared file volume (EFS / Filestore / Azure Files) mounted into the backend for storing agent working directories and generated artifacts | All components run inside a dedicated namespace in your cluster and are fully contained within your cloud account. DuploCloud does not operate any shared infrastructure on your behalf — your data stays in your environment. *** ## Access DuploCloud Requires To perform the installation, our team will need: * **Kubeconfig access** to your Kubernetes cluster (or a role with permissions to deploy Helm charts into the target namespace) * **Read access** to confirm the required infrastructure (load balancer, storage, certificates) is in place Access requirements are scoped to the installation namespace and can be revoked after deployment is complete. *** ## After Installation Once the deployment is complete, DuploCloud will provide: * The **application URL** where your team can log in * Confirmation that all services are healthy * Onboarding guidance for creating your first workspace, inviting users, and configuring AI agents For questions about the installation process, contact your DuploCloud account team.