Lilly Tech Systems
Kubernetes for AI (CKA + AI Focus)
Everything you need to pass the Certified Kubernetes Administrator exam with a focus on AI/ML workloads. Learn GPU scheduling, ML training jobs, model serving, networking for AI APIs, and persistent storage for datasets — all free. Includes 20+ scenario-based practice questions.
Your Study Path
Follow these lessons in order for complete exam preparation, or jump to any topic you need to review.
1. Exam Overview
CKA exam format (17 tasks, 2 hours), AI-specific skills, scoring, registration, environment setup, and what to expect on exam day.
2. Core Concepts
Pods, Deployments, Services, and ReplicaSets for ML workloads. Container fundamentals and resource management. Practice questions included.
3. GPU Scheduling
GPU resources, NVIDIA device plugins, node affinity, taints and tolerations for GPU nodes, and resource quotas. Practice questions included.
4. ML Workloads
Training jobs with Jobs and CronJobs, model serving with Deployments, Kubeflow operators, and distributed training. Practice questions included.
5. Networking & Storage
PersistentVolumeClaims for datasets, Ingress for model APIs, NetworkPolicies, and storage classes for ML pipelines. Practice questions included.
6. Practice Exam
20 scenario-based questions simulating real CKA exam tasks focused on AI/ML infrastructure. Detailed explanations for every answer.
7. Exam Tips
Cheat sheet, kubectl shortcuts, exam day strategy, time management, frequently asked questions, and additional resources.
What You'll Learn
By the end of this course, you will be ready to:
Pass the CKA Exam
Achieve the 66% score needed to earn your Certified Kubernetes Administrator certification with confidence in AI/ML scenarios.
Schedule GPU Workloads
Configure NVIDIA device plugins, set GPU resource requests and limits, and use node affinity to target GPU nodes for training jobs.
Deploy ML Models
Run training jobs with Kubernetes Jobs, deploy model serving endpoints with Deployments, and manage the full ML lifecycle on K8s.
Manage AI Infrastructure
Configure persistent storage for datasets, set up ingress for model APIs, and implement network policies for secure AI pipelines.