Red Hat at IBM TechXchange 2026

In this hands-on lab, attendees will build a Kubernetes-native Retrieval-Augmented Generation (RAG) application using enterprise data. Starting with unstructured documents, we will transform content into query-ready knowledge, create an ingestion and retrieval pipeline, and deploy the application on Kubernetes. Along the way, attendees will see how RAG can improve response quality without model fine-tuning while supporting governance, scalability, and data control. The lab concludes with a working AI-powered ticketing assistant that demonstrates practical patterns for building trusted, production-minded AI applications.

Red Hat® User Group (RHUG) events offer a unique opportunity for Red Hat technology users to gather and discover solutions to IT challenges with subject matter experts & peers. Join the next RHUG to network and explore cutting-edge solutions for IT modernization. Don’t miss this opportunity to transform your IT services and build lasting connections with industry professionals. Segment 1: IBM Fusion & Red Hat OpenShift Segment 2: IBM’s Client Zero Story Segment 3: IBM & Red Hat Lightwell Q&A (if time permits)

Terraform excels at Day-0 cloud infrastructure provisioning (Infrastructure-as-Code). For Day-1 configuration and application deployment, however, Red Hat Ansible Automation Platform is more suitable. While these tools have some overlapping capabilities, it's simpler to use Terraform for initial provisioning and Ansible Automation Platform for ongoing configuration and orchestration. Now, bring in Vault to provide automated privilege and secrets management, and you have one heck of a secure orchestration and automation solution for your business.

AI-powered vulnerability scanners are finding bugs faster than humans can fix them. Meanwhile, your production apps run on pinned dependency versions you can't upgrade without triggering a compliance nightmare. This is the new security gap: not detection, but remediation at machine speed. We'll walk through how teams running mission-critical workloads in highly regulated industries are using automation backports to remediate and secure the exact versions in production—no forced upgrades, no six-month test cycles. You'll see how cryptographic signing, policy-as-code, and AI-assisted exploit verification collapse the patch-to-production timeline from 90 days to under 10. Real architecture, real pipelines, real SLAs.

This hands-on lab is about deploying, managing, and securing agentic AI applications in a hybrid cloud using Red Hat OpenShift Platform Plus workflows. Topics include unifying infrastructure and deploying VMs, containers, and AI apps across multiple clusters using Red Hat Advanced Cluster Management (RHACM). Persistent storage for applications and artifact management is covered via Red Hat OpenShift Data Foundation and Red Hat Quay. Security is addressed with Red Hat Advanced Cluster Security (RHACS) for vulnerability scanning, network visualization, and guardrails. Finally, you'll implement multi-factor keyless workload authentication using Red Hat’s zero trust identity manager for access control to systems and data in an open, extensible sovereign foundation.

Manage the sprawl of static secrets like API keys, passwords, and SSH keys. Storing these credentials in playbooks or source control, even encrypted, increases risk and creates a compliance burden. We will share the tangible benefits achieved by centralizing secrets with HashiCorp Vault and plans to adopt new Ansible Automation Platform capabilities to achieve true workload identity. The session will then detail the "how," demonstrating the newly available technical integration between Ansible Automation Platform and HashiCorp Vault. Attendees will learn how new platform capabilities, such as AAP acting as an OpenID Connect (OIDC) provider for workload identity and Vault's dynamic secret engines, provide the ideal and auditable solution to complete this zero trust model.

This hands-on lab session offers a practical introduction to Model-as-a-Service (MaaS) with Red Hat AI, showing you how to deploy and manage generative AI models at scale to meet specific enterprise needs. These models can be served for business insights as well as linked to other AI agents to perceive the world around them and complete tasks. We'll cover deploying models on your own infrastructure, ensuring scalability and data sovereignty. You'll get a clear overview of integrated services, access controls, and how to view usage metrics for chargeback. We'll also explore practical, high-impact use cases like AI-powered code assistants for developers and agentic AI applications, demonstrating the immediate business value MaaS can deliver.

Enterprises adopting agentic AI face a gap: orchestration platforms lack sovereign runtime control, inference platforms lack multi-agent coordination. This session shows how watsonx Orchestrate and Red Hat AI bridge that gap with an open-standards architecture for enterprise agentic AI. We cover three layers. The agent runtime: Red Hat AI's Kubernetes-native inference with llm-d/vLLM and agent lifecycle management. The developer experience: GenAI Studio for model customization and context-aware agents via AutoRAG. Agentic orchestration: watsonx Orchestrate for multi-agent coordination with enterprise guardrails, plus Project Navigator using MCP to let agents manage the AI platform itself. Includes live demo and patterns for building secure, scalable agentic AI grounded in enterprise data

Last year, we shared our vision for generative and agentic AI. Today, that vision has evolved into a unified platform. IBM watsonx.ai and Red Hat AI Enterprise unite to deliver the predictive and generative capabilities needed to build, deploy, and manage AI at scale. Combined with IBM watsonx Orchestrate, this ecosystem provides a reliable, and scalable solution for running agentic AI in production. This session unveils the CY27 roadmap. We will cover: distributed inferencing with llm-d, scalable and auditable RAG, advanced model customization, multi-agent collaboration and advanced tool use, and security, governance, and trust. Get a first look at what’s new and next, and learn how to transition generative AI and agents from experimentation to production quickly and reliably.

Lightwell safeguards use of open source software in application development. In this deep-dive session, we map out a comprehensive technical roadmap and architectural framework in Lightwell - which uses a generative AI-powered remediation engine that combines frontier and open AI models with human engineering expertise. Identify, validate, and patch vulnerabilities early across critical dependencies embedded in specific software packages running in production. Examine topics from core platform necessity, backported remediation and policy guardrails to deployment considerations and post-onboarding optimization strategies. Leave with actionable insights, technical clarity, and a product strategy that delivers advanced, AI-driven automated pipelines at scale in a hybrid cloud environment.

In this interactive lab, you’ll learn how to position Red Hat Ansible Automation Platform as the execution layer for agentic IT operations. We’ll move beyond static playbooks to showcase how autonomous workflows built on Ansible Automation Platform enable context-aware decision-making, adaptive remediation, and policy-driven governance across hybrid environments—including Red Hat Enterprise Linux, Windows, cloud, and network infrastructure. You’ll explore how agents consume observability signals, reason using large language models, and execute actions—all while maintaining human oversight and auditability. Attendees will design and deploy end-to-end agentic workflows that interpret alerts, gather context, and trigger self-healing automation in real time.

Automated remediation handles known problems well. But when an incident could match multiple runbooks, or needs investigation before action, today's pipelines hit a wall: they weren't designed to reason about which response fits best. In this session, IBM and Red Hat show how Instana and Ansible Automation Platform (including Event-Driven Ansible), and AI agents via MCP move from "run this playbook" to choosing the right action per incident. Agents interpret telemetry, investigate causes, compare automation options, and select responses. Red Hat Ansible executes with security controls including RBAC, auditability, reproducibility, and human-in-the-loop approval. Leave with practical patterns, lessons learned from production, and a clear path from insight to decision to governed action.

This session provides an in-depth look at how persistent storage is designed and implemented in OpenShift Virtualization environments, with a focus on architecture patterns, configuration strategies, availability architecture, and performance tuning. Learn how to select and configure container storage interface (CSI) drivers, design storage classes for virtual machine (VM) workloads, using platform features for efficient provisioning and import workflows. Explore live migration requirements, snapshot and clone capabilities, to align storage design with high availability and disaster recovery objectives. Leave with a clear understanding of how to build and operate a reliable, high-performance storage foundation for data consistency, scalability, and operational efficiency in production.

Patch a CVE in under an hour, start to finish, with full compliance artifacts. You'll configure a Lightwell backport feed for your artifact repository. Write Ansible tasks that detect vulnerable JARs in your build, swap in the patched version, and rebuild the container. Add a step that cryptographically signs the result and uploads the SBOM. Wire it into a Jenkins job. Run it. Watch it fail because you forgot to set credentials. Fix that. Run it again. Watch it work. Then we'll throw a second CVE at it and confirm the pipeline self-heals. Pair programming is encouraged but not required. What you need: laptop, SSH client, and some Ansible experience. What you'll build: the whole remediation loop, ready to clone into your own GitLab.

Modern DevOps teams are under constant pressure to meet strict SLAs while managing increasingly complex, distributed systems. In this hands-on lab, participants will design and deploy an agentic AI-powered DevOps assistant using Red Hat OpenShift AI on IBM Cloud. This intelligent assistant automates key aspects of incident management—detecting anomalies in logs and telemetry, diagnosing root causes, and proposing actionable remediation steps. By combining AI-driven reasoning with enterprise-grade DevOps workflows, the lab demonstrates how agentic systems can augment engineering teams and significantly reduce mean time to repair (MTTR). Participants will build end-to-end workflows that integrate with monitoring tools, CI/CD systems, and ticketing platforms

Frontier AI models now discover zero-days faster than teams can patch them—your window to respond has collapsed from weeks to hours. But upgrading legacy apps for CVEs means six weeks of regression testing you don't have. This session shows how regulated industries use version-specific backports and automated supply chain controls to patch legacy stacks without breaking production. You'll see the actual workflow: from CVE to cryptographically signed fix to deployed container - with compliance built in. No vendor fluff—just the architecture and automation that makes it work. From prioritizing critical threats with IBM Concert to your trust execution layer for patching and orchestration workflows with Red Hat Ansible Automation Platform, learn to build resilience from app to infrastructure.