Red Hat OpenShift AI
MLOps & GenAIOps (AI267)
Bridge the gap between data science and IT operations. Learn to train, optimize, serve, and monitor intelligent applications—including Large Language Models (LLMs)—at scale. Prepare for the Red Hat Certified Developer in AI (EX267) exam.
GenAI & MLOps
Master the complete lifecycle of both predictive machine learning and Generative AI models.
Kubeflow Pipelines
Automate ML workflows and data science pipelines using Kubeflow SDK and Elyra.
LLMs & RAG
Deploy, optimize, and evaluate Large Language Models (LLMs) and build RAG applications.
TrustyAI Monitoring
Monitor models in production for bias and data drift to ensure ethical AI performance.
Why OpenShift AI (AI267)?
Artificial Intelligence is moving out of the lab and into production. Organizations need platforms that can standardize the chaos of machine learning workflows. Red Hat OpenShift AI provides a complete MLOps and GenAIOps platform to efficiently train, test, deploy, and monitor models at scale.
The AI267 course teaches you how to operationalize the AI lifecycle. You will move beyond basic model serving to build production-ready GenAI applications using RAG patterns, vLLM runtimes, and Kubeflow pipelines.
Prerequisites: Python programming experience (AD141) and OpenShift basics (DO288) are highly recommended. A basic understanding of machine learning and LLM concepts is required.
Your Learning To Placement Journey
-
1
AI/ML Fundamentals
Set up Jupyter notebooks, data science projects, and connect external data sources.
-
2
Model Serving & Pipelines
Deploy models via KServe, monitor drift with TrustyAI, and automate with Kubeflow.
-
3
EX267 Exam Prep
Performance-based lab practices to clear the Red Hat Certified Developer in AI exam.
-
4
Placement Assistance
Direct pipeline to top enterprises hiring MLOps Engineers and AI Platform Architects.
Official Course Syllabus (AI267)
01
Introduction to Red Hat OpenShift AI
- Understand OpenShift AI architecture and MLOps/GenAIOps concepts
- Identify how OpenShift AI integrates with OpenShift Container Platform
- Configure collaborative data science projects
- Manage user access and specialized hardware resources
02
AI/ML Development Workbenches
- Deploy and manage Jupyter Notebook environments
- Create custom workbench images for specialized AI workloads
- Configure data connections to external storage (S3, Databases)
- Monitor resource usage with TensorBoard
03
Deploy & Serve Predictive Models
- Understand model serving workflows and KServe architecture
- Deploy models using Standard and Advanced serving modes
- Serve predictive AI models using the OpenVINO runtime
- Store and retrieve artifacts from OCI containers and PVCs
04
Monitor AI Models in Production
- Configure model observability and metrics
- Monitor deployed models for data drift using TrustyAI
- Detect and mitigate model bias in real-time
- Analyze hardware consumption with Grafana and Prometheus
05
Data Science Pipeline Automation
- Create and manage AI pipelines using the Elyra visual builder
- Develop advanced pipelines using the Kubeflow SDK
- Implement container components and artifacts management
- Execute systematic experimentation for production workflows
06
GenAI Model Optimization & Evaluation
- Select and import models from the RHOAI catalog and Hugging Face
- Serve Large Language Models (LLMs) using the vLLM runtime
- Optimize models using LLM Compressor (quantization techniques)
- Evaluate LLM performance using standard benchmarks (LMEval)
07
Building GenAI Applications
- Understand industry patterns for GenAI and trustworthy AI practices
- Build Retrieval-Augmented Generation (RAG) applications
- Integrate vector databases and document processing
- Implement safety guardrails and agentic workflows
Pricing varies based on online/offline & combo packages.