Learning
Self-paced tracks for hands-on engineers. Each track is a sequence of numbered modules with diagrams, code, and exercises. Pick a track below to see its modules.
9 tracks · 142 modules total.
ACM Multicluster
Operate ten or fifty OpenShift clusters from one control plane. Red Hat Advanced Cluster Management end-to-end — registration, policies, GitOps, applications, observability, hosted control planes, and disaster recovery.
Agentic AI
Build agents that loop, use tools, and produce results. From the agent loop and MCP to planning, evaluation, and production.
Cisco APIC
Learn APIC policy, tenants, EPGs, contracts, access policies, and L3Out concepts before touching a real ACI fabric.
Cloudflare
The unified network + edge + security + AI platform. Tunnels, Zero Trust, Magic networking, Workers, Pages, and the rest.
DevSecOps
Shift security left into the SDLC, automate the gates at every stage, and treat security as a property of every commit. Threat modeling, SAST/SCA, container and supply-chain security (SLSA, SBOM, cosign), IaC scanning, runtime detection (RHACS, Falco), secrets, zero trust, compliance, and SIEM — ending in a fully gated deployment pipeline.
Hands-On Open Liberty
End-to-end application development with Open Liberty. Provision a dev VM, install the toolchain, write a Jakarta EE 10 / MicroProfile 6.1 service, then add the real-world stack on top — PostgreSQL, Redis, Kafka, WSO2 MI/IS/APIM, and SigNoz observability — running as a fleet of containers on one VM. The classroom track for backend engineers before they touch production.
Kubeflow
The platform layer for the ML workflow on Kubernetes. Upstream Kubeflow end-to-end — notebooks, KFP pipelines, distributed training, Katib HPO, KServe model serving, multi-tenancy with Profiles, and production-grade install patterns.
OpenShift: Disconnected Deployment
Stand up OpenShift in an environment that cannot reach the public internet. The four supply chains, a Quay-backed mirror, oc-mirror v2 workflow, agent-based install, the Day-1 baseline (IDMS/ITMS/CatalogSources), GitOps handoff, operator pinning, and the Day-2 mirror operations that keep it healthy. Anchored to the greenfield v7 fleet running on dl385-2.
Practical Data Science
End-to-end data science on a real on-prem GPU server. Stand up a multi-user ML platform — JupyterHub, MLflow, MinIO, Slurm, Prefect, dbt, vLLM, Label Studio, Prometheus — then use it to ship four capstones: tabular ML, a modeled warehouse with dashboards, a self-hosted RAG application, and a multi-GPU computer-vision model. 2× NVIDIA L40S included.