Red Hat Summit 2025

Docling, an open source package, is rapidly becoming the de facto standard for document parsing and export in the Python community. Earning close to 8,000 GitHub stars in just 10 days, Docling is redefining document AI with its ease and speed of use. In this session, we’ll introduce Docling and its features.

Red Hat, IBM, and Intel’s partnership has created opportunities to build and deploy architectures that accelerate innovation for several use cases, including:

• Artificial intelligence (AI) and machine learning (ML) and large language model (LLM) workloads.
• Auto-healing Red Hat OpenShift clusters.
• Rapid deployment of Red Hat OpenShift with IBM Fusion.
• Multisite resilience.
• Consolidating containers and virtual machines (VMs).

Join this panel discussion for an overview of the college’s Red Hat OpenShift and IBM watsonX Fusion project

In today’s fast-paced and unpredictable world, resilience and adaptability are essential for success. Join us for a dynamic panel discussion with industry leaders who have successfully navigated uncertainty, embraced change, and built resilience within their organizations and teams.   Our expert panelists will share strategies, lessons learned, and actionable insights to help you strengthen your adaptability–professionally and personally. Whether you’re a seasoned leader or new in your career, this conversation will offer valuable perspectives on how to stay agile, foster innovation, and build a culture of resilience.

Bring your curiosity, questions, and ideas to this thought-provoking discussion and walk away with fresh perspectives to help you navigate and thrive in an ever-changing world.   Stick around after the session to network with panelists and attendees with light refreshments from 4:30 to 5:30pm.

Panelists:

• Marisa Porges, Chief of Staff, IBM
• Alexandra Schoen, Automation Solutions Leader, World Wide Technology (WWT)
• Jen Bailin, Vice President, North America, West and Central Region, Red Hat
• Gabie Boko, Chief Marketing Officer, NetApp
• Constanza Cabello, Managing Director, Global Inclusion & Diversity, State Street Corporation

AI models rarely operate in isolation—they span public cloud, on-prem, and edge compute environments. This creates challenges in workload mobility, security, and data control. Fragmented infrastructures slow down AI pipelines, introduce silos, and increase the complexity of scaling AI models across environments.   This session focuses on building an intentionally-designed hybrid cloud infrastructure that adapts to dynamic workloads. Learn how to:

• support AI model training and deployment with highly secure, high-performance computing;
• make application modernization simpler across hybrid cloud environments;
• deploy storage solutions that enhance AI workloads with flash-optimized architecture and intelligent data placement.   The result is a unified AI infrastructure that scales horizontally and vertically—running AI models where the data is.

This talk explores our journey of migrating 70% of our platform from VMWare to Red Hat OpenShift Virtualization. We’ll discuss the challenges, successes, and best practices that helped us achieve consistent scalability, flexibility, and performance while maintaining a cohesive infrastructure.  We’ll also examine the experience of designing and scaling an internal platform based on OpenShift Virtualization, including:

• Our decision-making process and the key benefits of using OpenShift Virtualization over other solutions, especially for scaling. 
• How we use the platform to support GPU-enabled AI workloads with Red Hat AI, Red Hat OpenShift AI, and IBM watsonx at scale. 
• Key architectural considerations―from networking and storage design to infrastructure planning―that let us scale out the platform efficiently. 
• Implementing Red Hat OpenShift GitOps methodology for Red Hat OpenShift and VM lifecycle management, security policies, and integration with CI/CD pipelines to streamline operations.
• Techniques and tools for monitoring, optimizing, and troubleshooting a high-volume OpenShift Virtualization environment with IBM Instana Observability, IBM Turbonomic, and Red Hat Advanced Cluster Management. 
• Lessons learned from scaling, including handling resource contention, achieving consistent VM performance, and balancing VM and container resources.

Speakers:

• Dmitry Zhukovski, Principal Solution Architect, IBM
• Jason Erickson, Principal Technology Zone Manager, IBM

Enable your AI models to perform better on specialized tasks while retaining control of your data. Discover how easy it is to fine-tune LLMs using taxonomy-driven data curation and synthetic data generation, in the training and deployment of AI models that's tailored to your domain. Join us in taking your first steps to simplify model tuning at scale in the cloud, while keeping your costs and complexity low.

Speaker: Jason McGee, IBM

AI innovation is often constrained by rigid infrastructure and fragmented pipelines. As AI models grow more complex, training, deployment, and orchestration processes become disjointed, slowing the entire lifecycle. An open AI ecosystem can remove these bottlenecks and speed model-to-production workflows. In this session, we’ll share technical demonstrations of how to unite isolated AI into flexible, cross-environment pipelines that simplify model training, tuning, and deployment across cloud and on-premise environments. Join us to learn how an adaptive AI platform breaks down barriers to foster innovation and achieve efficiency and security at scale.

Speakers:

• Suhas Kashyap, Sr. Product Manager, watsonx.ai, IBM
• Emma Gauthier, Product Marketing Manager, IBM

Join this session to learn how to kick-start generative AI solutions with Red Hat AI Instruct Lab on IBM cloud. With limited data science expertise, train a model using InstructLabs' intuitive user interface and use its synthetic data generation features to create thousands of data points that quickly fine tune large language models (LLMs).

Speakers:

• Younes Ben Brahim, AI BU, Red Hat
• Dan Waugh, Principle Product Manager, Red Hat AI InstructLab, IBM

The National Energy Systems Operator (NESO) is the unseen force that powers Great Britain—ensuring households, businesses, and public services stay connected to a reliable electricity supply. Behind this is a sophisticated balancing system that keeps energy consumption in perfect harmony with production. Yet the legacy system—designed in the 1990s when fossil fuels were king—was ill-equipped to manage the modern grid, where energy comes from countless decentralized sources like wind turbines, solar panels, and batteries. In response, NESO embarked on an ambitious transformation journey to overhaul its grid-balancing solution. The result? An innovative Open Balancing Platform (OBP) went live in December 2023, powered by Red Hat OpenShift Platform Plus, streams for Apache Kafka, and Red Hat OpenShift Service Mesh. In this session, you’ll discover how NESO used Red Hat OpenShift and associated technologies like Red Hat OpenShift Pipelines and Red Hat OpenShift GitOps to radically accelerate the delivery of business value. With enhanced security and unmatched resilience, NESO’s development and operations teams are now able to deploy frequent, incremental updates without compromising on safety or uptime. Learn how they harnessed the power of hybrid cloud, using Microsoft Azure to test and develop rapidly, while delivering changes to core UK Critical National Infrastructure every sprint—without any downtime during releases. This transformative project wasn’t just a success; it earned recognition at the 2024 Computing.co.uk DevOps Excellence Awards, where the OBP won Best DevOps Team of the Year.

Speakers:

• Nisha Bhamidimarri, , Product Manager, National Energy System Operator
• Matthew Peacock, , Senior IT Architect (Lead Operational Architect), IBM
• Jon Walton, Solution Architect, Red Hat
• Jordan Sherne, IT Architect, National Energy System Operator

This session explores how Red Hat OpenShift on IBM Cloud helped the Brazilian government bring essential financial services to millions of citizens through a mobile app. Learn how Red Hat OpenShift’s scalability and flexibility helped create financial accessibility while delivering critical services, efficiently and with security. Participants will learn about the architecture and key strategies that allowed the state-owned bank to meet demand while maintaining high standards of security, performance, and compliance.

Speaker: Olavo Borges, Senior Cloud Architect, IBM

AI adoption often stalls—not because of AI itself—but because of IT bottlenecks. Cloud provisioning is slow, app deployments are complex, and manual workflows create a reactive environment. As AI models evolve, these outdated processes can’t keep up, delaying innovation and making AI scaling a cumbersome task.   This session explores how AI can enhance automation to transform IT operations into self-optimizing systems. Demos will highlight the latest integrations with Red Hat solutions that create dynamic resource allocation, automated workflows, and intelligent monitoring—empowering customers to build an IT environment ready for AI innovation.

Speakers:

• Matt Rodkey, Program Director - watsonx Code Assistant for Ansible Software, IBM
• Daniel Savage, Senior Director of Product Management, HashiCorp
• William Lobig, Vice President, IBM Automation, Product Management, IBM

In a pioneering journey toward artificial intelligence (AI) model efficiency, Red Hat and IBM Research joined forces to tackle one of AI's most pressing challenges: the high resource demands of large model fine-tuning. Faced with the need to balance scale with cost-efficiency, we developed an advanced open source tuning stack powered by PyTorch Fully Sharded Data Parallel (FSDP), Hugging Face, and the Kubeflow Training Operator (KFTO), all within the FMS HF Tuning utility. Scaling this technology across AI teams posed some initial challenges, but through a collaborative, community-focused approach, we created a blueprint for AI and machine learning (ML) practitioners to leverage the power of large models on minimal infrastructure. In this session, we’ll walk you through our journey and share actionable insights, including:

• Building an open, collaborative foundation that drives efficient model tuning practices
• Approaches to overcoming adoption barriers and enabling user-friendly customization
• Techniques for integrating LoRA and quantized LoRA for high-throughput, low-memory model fine-tuning
• A live demonstration fine-tuning a 405B parameter LLaMA model on a single-node Red Hat OpenShift AI cluster Join us to discover how OpenShift AI’s tuning stack can transform model customization across industries, helping your teams harness the power of AI with minimal resources and maximum impact.

Speakers:

• Praveen Jayachandran, Senior Technical Staff Member, AI Platform, IBM
• Amita Sharma, Engineering Manager, Red Hat

Planned for eventual sunsetting, AppAlpha—a monolithic, three-tier app built with IBM WebSphere Application Server and Database—required too much time and effort to redesign and rewrite for modern use. In this session, we’ll share how IBM’s CIO organization used this application to complete a proof of concept on migration of legacy applications from a private cloud environment to Red Hat OpenShift Virtualization. Key topics include:

• Systematic assessment of the existing architecture to develop a technical solution with minimal changes, using the standard configuration and automation with Red Hat Ansible Automation Platform.
• Creating a virtual machine (VM) in an OpenShift Virtualization environment, including adding persistent volumes for installing middleware and improving performance for IBM Db2 transactions with IBM Power Systems.
• Evaluating migration toolkit for virtualization, Red Hat’s suggested solution, versus a lift-and-shift approach to application deployment to a VM.

Speaker: Sanjeev Kumar Marimekala, Thought Leader, STSM and Enterprise Architect, IBM

IBM's CIO is using the power of Red Hat’s comprehensive portfolio to drive digital transformation and enhance IT operations for hundreds of users. By adopting a strategic approach to automation and security, the Automation Tools team achieved significant improvements in efficiency, scalability, and compliance.  In this session, the team will share their experiences with: 

• Red Hat Ansible Automation Platform as the foundation. Discover how they streamlined infrastructure provisioning, configuration management, and application deployment. 
• Event-Driven Ansible. Learn how they automated complex workflows triggered by real-time events to monitor and remediate systems, improving response times and reducing manual intervention. 
• Red Hat Single Sign-On (RHSSO) to Keycloak Migration: Discover the challenges faced and the strategies employed to ensure a seamless transition to a more robust identity and access management solution. 
• Advanced Ansible Automation Platform configurations. Explore how the team extended the capabilities of Ansible Automation Platform to meet their unique business requirements, including integrating with external cloud databases.  By sharing these insights, this session aims to provide valuable guidance on how organizations can apply similar techniques to unlock the full potential of Red Hat solutions.  Speaker: Jorge Senger, Hybrid Cloud Architect, IBM

IBM and Red Hat are working to take artificial intelligence (AI)—including generative AI (gen AI) and foundation models—everywhere, including to edge-based deployments where data gravity and reduced resources and expertise are a reality. In this session, to demonstrate the capabilities of the Red Hat Edge hub and spoke (edge to cloud) deployment and continuous development stack, we'll show demonstrations from industry 4.0, 5G analytics, and smart infrastructure threat monitoring. Our demonstrations will include:

• Example deployments on canonical hardware/software (HW/SW) integrated architectures (e.g. a three server edge spoke cluster)
• Benchmark results of the time series foundation model library for industry 4.0
• Novel work on an AI accelerator from IBM Research, supported by Red Hat

Speakers:

• Alberto Valdez Garcia, Principal Research Scientist, Manager, IBM
• Myrian Fentanes, Product Manager, Red Hat
• Catherine Crawford, IBM Fellow, IBM