Enterprise AI Infrastructure WorkShop

Optimize, scale, and simplify your enterprise AI workloads with a fully customizable, hands-on learning experience.

More information
Jump to section

The Enterprise AI Infrastructure Workshop is a 24-hour, practical engagement designed to help organizations maximize the value of their Red Hat OpenShift AI (RHOAI) investment.

This is a flexible, 5-module modular workshop. You can tailor the learning path to your organization’s specific infrastructure maturity, goals, and technical needs.

Duration:
24 hours (typically spread across 2–3 days)

Format:
Hands-on, activity-driven labs and practical scenarios

Flexibility:
Pick and choose modules or run the complete end-to-end curriculum

What You Will Achieve

  • Reduce Adoption Risk: Build a robust, enterprise-ready AI architecture from day one.
  • Maximize ROI: Optimize expensive GPU investments using advanced sharing and slicing techniques.
  • Accelerate Delivery: Eliminate infrastructure bottlenecks so developers and data scientists can iterate faster.

Who Should Attend?

This workshop is designed to bridge the gap between platform infrastructure and AI/ML teams:

  • Platform Engineers & Cluster Administrators: Learn how to architect, scale, govern, and optimize the underlying AI infrastructure.
  • Data Scientists & Developers: Gain practical insights into leveraging GPU acceleration, automated scaling, and Model-as-a-Service (MaaS) workflows.

Tailor Your Learning Experience

Every enterprise AI journey is unique. During our initial scoping conversation, we’ll work with your team to adapt and prioritize these five core modules based on your current infrastructure maturity and goals.

OpenShift AI Installation & Configuration

Module 1: OpenShift AI Installation & Configuration

Objective: Master the deployment lifecycle and fundamental setup of Red Hat OpenShift AI.

  • OpenShift AI Architecture: Overview and deployment lifecycle
  • Installation: Step-by-step installation and base setup
  • Component Configuration: Customizing core RHOAI components for production
 Accelerator Integration & GPU-as-a-Service

Module 2: Accelerator Integration & GPU-as-a-Service (GPUaaS)

Objective: Maximize hardware efficiency and enable shared GPU infrastructure for training and inference.

  • Working with Hardware Accelerators: Integrating GPUs into your OpenShift cluster
  • GPU Optimization: GPU slicing and Multi-Instance GPU (MIG) setup
  • GPU-as-a-Service: Delivering shared acceleration capabilities across teams
 Production Training at Scale

Module 3: Production Training at Scale

Objective: Orchestrate, schedule, and run distributed AI training workloads efficiently.

  • Distributed Workloads Architecture: Structuring large-scale AI training
  • Intelligent Scheduling: Utilizing Red Hat Build of Kueue (RHBoK) for workload queuing
  • Workload Orchestration: Managing Ray-based workloads and Kubeflow training pipelines
  • Real-World Use Cases: Best practices for production training scenarios
 Production Inference & Model Serving

Module 4: Production Inference & Model Serving

Objective: Deploy, serve, and benchmark predictive and Generative AI models efficiently.

  • Fundamentals of Model Serving: Setting up scalable model endpoints
  • Generative & Predictive AI: Serving classical models alongside LLMs
  • Distributed LLM Inference (llm-d): Scaling large language model serving
  • Performance Benchmarking: Optimizing latency, throughput, and resource utilization
 Enterprise Model-as-a-Service

Module 5: Enterprise Model-as-a-Service (MaaS)

Objective: Establish enterprise-grade governance, self-service access, and cost management for AI APIs.

  • Governance & Security: Centralized AI security policies and access controls
  • Developer Self-Service: Enabling instant access to AI endpoints for application teams
  • AI Gateway: Leveraging Kubernetes-native routing and rate limiting
  • Cost Visibility: Tracking usage, cost attribution, and resource consumption

Get in touch with us

We don't believe in one-size-fits-all training. Leave your contact info below, and we'll connect to discuss your current OpenShift AI setup, target outcomes, and how to best blend these 5 modules for maximum ROI.