Uploaded image for project: 'OpenShift Container Platform (OCP) Strategy'
  1. OpenShift Container Platform (OCP) Strategy
  2. OCPSTRAT-2648

AI-assisted installation experience - Phase II

XMLWordPrintable

    • Icon: Outcome Outcome
    • Resolution: Unresolved
    • Icon: Critical Critical
    • None
    • None
    • None
    • Product / Portfolio Work
    • 17% To Do, 83% In Progress, 0% Done
    • False
    • Hide

      None

      Show
      None
    • False
    • None

      Outcome Overview

      OpenShift installation will be transformed from a fragmented, expertise-dependent process into a unified, AI-guided experience accessible through a single entry point.

      Users will no longer need to navigate the complexity of choosing between IPI, UPI, Agent-based, or other installation methods, nor will they need to search through hundreds of pages of scattered documentation. Instead, an AI-powered chat assistant at console.redhat.com will intelligently guide them to the appropriate installation approach based on their specific requirements, provide the necessary configuration files and artifacts, offer real-time troubleshooting support, and handle both connected and disconnected deployment scenarios.

      This represents tangible movement toward Red Hat's strategic goals of reducing installation friction, expanding OpenShift adoption, and delivering AI-powered experiences that fundamentally improve the customer journey.

       

      Success Criteria

      • Unified Entry Point: All OpenShift installation journeys begin at console.redhat.com through a single AI-assisted chat interface, regardless of target platform or connectivity scenario
      • Intelligent Installation Method Selection: The chat assistant successfully identifies and recommends the appropriate installation method (Assisted Installer, openshift-install, Agent-based, etc.) based on user requirements without requiring users to have prior knowledge of these options
      • Configuration Automation: The system generates installation configuration files (manifests) and artifacts (ISOs) tailored to user-specified requirements through conversational interaction
      • RAG-Enabled Guidance: In-context installation guidance is provided throughout the workflow, pulling from OpenShift documentation and best practices through Retrieval-Augmented Generation
      • Troubleshooting Capability: Users can paste installation logs into the chat for AI-powered analysis, receiving specific remediation steps and recommendations
      • Connected Deployment Support:
        • Assisted Installer-supported platforms (baremetal, vSphere) provide step-by-step guidance with live progress tracking
        • Non-Assisted Installer platforms (AWS, etc.) receive appropriate tool links and configuration files
      • Disconnected Deployment Support: Self-contained bootable ISOs are available for air-gapped environments, pre-packaged with RHCOS, core payload, pre-selected Day-2 operators, and installation UI for common use cases (OpenShift Virtualization, OpenShift AI, vanilla OpenShift with popular operators)
      • Day-2 Operations: Disconnected ISOs support upgrade operations and adding new nodes
      • User Validation: Early adopters and beta users confirm the experience significantly reduces installation complexity and time-to-value compared to existing processes

       

      Expected Results (what, how, when)

      Reduced Installation Friction and Time-to-Deploy

      • What: Measure reduction in average time from "decide to install OpenShift" to "running cluster"
      • How:
        • Track installation completion times through console.redhat.com telemetry (connected deployments)
        • Survey users comparing their experience to previous installations
        • Monitor support case volume related to installation confusion/failures

      Expanded Installation Method Accessibility

      • What: Measure adoption of previously underutilized installation methods that are now AI-recommended based on use case fit
      • How:
        • Track distribution of installation methods chosen through the AI assistant
        • Compare to historical distribution of installation methods
        • Identify patterns where AI recommends Assisted Installer, Agent-based and other approaches for scenarios where users previously defaulted to more complex methods

      Improved First-Time Success Rate

      • What: Measure the percentage of installations that complete successfully without requiring troubleshooting intervention or restart
      • How:
        • Track the completion status of installations initiated through the AI assistant
        • Monitor AI troubleshooting interactions (when users paste logs) and resolution outcomes
        • Survey users on whether they needed external support beyond the AI assistant

      Disconnected Deployment Simplification

      • What: Measure the adoption and success of the self-contained ISO approach for air-gapped environments
      • How:
        • Track download metrics for pre-generated ISOs by use case (OpenShift Virtualization, OpenShift AI, vanilla)
        • Survey disconnected deployment users on their experience compared to previous oc-mirror and similar workflows

      Documentation Engagement Shift

      • What: Measure reduction in direct documentation page visits as users rely on RAG-enabled in-context guidance
      • How:
        • Compare documentation site traffic patterns before and after AI-assisted installation launch
        • Track RAG query success rates and user satisfaction with provided guidance
        • Monitor documentation-related support ticket volume

       

      Post Completion Review – Actual Results

      After completing the work (as determined by the "when" in Expected Results above), list the actual results observed / measured during Post Completion review(s).

              mak.redhat.com Marcos Entenza Garcia
              mak.redhat.com Marcos Entenza Garcia
              Ju Lim, Michal Zasepa, Oved Ourfali, Xiaoli Tian, Zane Bitter
              Zane Bitter Zane Bitter
              None
              None
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated: