Practice for the Red Hat Certified Specialist in OpenShift AI (EX267) exam with 18 exam-style practice questions, instant answer reveals, and concise explanations of every correct answer. Topics include: Which component provides the visual frontend user interface for creating projects, launching workbenches, and monitoring. Follow @CertPunch and visit certpunch.com for more certification practice exams and study content.
Prefer hands-on? Take this round as an interactive practice test — answer every question, get instant feedback, and see your score: Start the Red Hat Certified Specialist in OpenShift AI (EX267) practice test →
What you will practice
- Which component provides the visual frontend user interface for creating projects, launching workbenches, and…
- When sharing a Jupyter notebook with a teammate via Git in OpenShift AI, what is the best practice to protect…
- Which framework can be embedded directly into custom notebook images to allow seamless Git integration direct…
- Which protocol is used by default by ModelMesh to communicate internally between the router and individual mo…
- How do you expose a deployed machine learning model endpoint securely to external client applications?
- What criteria does OpenShift AI evaluate to determine if a notebook workspace container is completely unhealt…
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. Which component provides the visual frontend user interface for creating projects, launching workbenches, and monitoring serving runtimes?
Answer: B. OpenShift AI Dashboard (provided by the odh-dashboard component)
The OpenShift AI Dashboard, powered by the odh-dashboard component, provides the graphical interface for managing projects, workbenches, and model serving. Candidates should recognize this as the primary UI.
Q2. When sharing a Jupyter notebook with a teammate via Git in OpenShift AI, what is the best practice to protect secrets?
Answer: B. Use environment variables backed by an OpenShift Secret or Data Connection
Using environment variables backed by OpenShift Secrets or Data Connections keeps credentials secure when sharing notebooks. Hardcoding credentials directly inside cells exposes them in Git history.
Q3. Which framework can be embedded directly into custom notebook images to allow seamless Git integration directly from the UI toolbar?
Answer: C. JupyterLab Git extension (jupyterlab-git)
The JupyterLab Git extension allows seamless Git integration directly from the notebook interface toolbar. This is pre-installed in standard OpenShift AI workbench images for committing code.
Q4. Which protocol is used by default by ModelMesh to communicate internally between the router and individual model runtimes?
Answer: D. gRPC
ModelMesh uses gRPC by default for efficient internal communication between the router and individual model runtimes. Expect questions on how runtime components exchange data and scale.
Q5. How do you expose a deployed machine learning model endpoint securely to external client applications?
Answer: D. Through an OpenShift Route or Ingress endpoint created by the service
Deployed machine learning models are exposed to external clients securely via standard OpenShift networking resources. Know that Routes or Ingress endpoints provide this external access.
Q6. What criteria does OpenShift AI evaluate to determine if a notebook workspace container is completely unhealthy?
Answer: A. Kubernetes Liveness and Readiness probes
OpenShift AI relies on standard Kubernetes Liveness and Readiness probes to determine if a notebook workspace container is unhealthy. This is a fundamental Kubernetes concept you must understand.
Q7. Which Python SDK is used to programmatically build and compile pipelines compatible with OpenShift AI?
Answer: A. KFP (Kubeflow Pipelines) SDK
The KFP (Kubeflow Pipelines) SDK is used to programmatically build and compile pipelines compatible with OpenShift AI. Be aware that OpenShift AI uses Tekton pipelines under the hood.
Q8. Where inside the dashboard UI can you view the historical execution durations for distinct data science pipeline instances?
Answer: A. Pipeline Runs history tracking logs view
Historical execution durations for pipeline instances are viewed in the Pipeline Runs history tracking logs view. Navigate to the Pipelines section to track run metrics.
Q9. Which action allows an AI model to continue accepting external calls when a physical worker node experiences failure?
Answer: C. Setting replica counts greater than one per model
Setting replica counts greater than one allows a model to continue accepting calls if a worker node fails. The Kubernetes scheduler automatically restarts pods on healthy nodes.
Q10. What file extension is generated when exporting a compiled data science pipeline for import into OpenShift AI?
Answer: D. .yaml or .zip file format
Compiled data science pipelines are exported in a compressed YAML file format, typically as a .yaml or .zip file. You will import these pipeline definitions directly via the dashboard.
Q11. How can a data scientist execute a pipeline run automatically at regular intervals?
Answer: B. Configure a Scheduled Run or Cron trigger in the pipelines interface
Configuring a scheduled run or recurring trigger in the pipelines UI is the standard way to automate executions. Leaving a notebook running continuously or writing external bash scripts are unsupported hacks that fail to leverage native pipeline orchestration features.
Q12. To inject environment variables securely into an OpenShift AI workbench, which native OpenShift resource should be linked?
Answer: C. Secret
Secrets are the standard OpenShift resource designed to securely store and inject sensitive environment variables into workbenches. ConfigMaps are useful for non-sensitive configuration data but fail to protect credentials or API keys like a Secret does.
Q13. Which operator must be present on your OpenShift Container Platform to orchestrate accelerated NVIDIA GPU hardware configurations?
Answer: C. NVIDIA GPU Operator
The NVIDIA GPU Operator handles provisioning the necessary drivers and runtime configurations for GPU acceleration. The Node Feature Discovery Operator detects hardware capabilities but does not configure or orchestrate the actual GPU software stack.
Q14. Where can you view the status of all pipeline runs within an active Data Science Project?
Answer: D. Pipeline section under the project dashboard user interface
The Pipelines section of the Data Science Project dashboard provides a centralized UI to view all active and historical pipeline runs. Terminal logs or operator installation screens fail to provide aggregated run tracking for your data science projects.
Q15. Which runtime is commonly used for serving deep learning models inside the ModelMesh framework?
Answer: C. Triton Inference Server platform
The correct answer is correct because the Triton Inference Server is the standard runtime used within ModelMesh to serve complex deep learning frameworks efficiently. It specifically handles multi-framework model serving, unlike generic web or Node.js servers.
Q16. What information is contained within an OpenShift AI accelerator profile?
Answer: C. Hardware labels and tolerations matching specific accelerator nodes
The correct answer is correct because an accelerator profile configures tolerations and identifies hardware labels so workbenches and runtimes can target specialized nodes. This ensures workloads schedule correctly on hardware like GPUs.
Q17. Which OpenShift storage type is typically utilized for a workbench's persistent storage?
Answer: D. ReadWriteOnce (RWO) Persistent Volume Claim
OpenShift AI workbenches require a ReadWriteOnce Persistent Volume Claim to retain user files and notebook states across pod restarts. HostPath and EmptyDir volumes do not provide the necessary data persistence for reliable workbench operations.
Q18. What component ensures that the OpenShift AI environment matches state definitions kept in code repositories?
Answer: C. Red Hat OpenShift GitOps Operator application
The Red Hat OpenShift GitOps operator utilizes ArgoCD to continuously reconcile the live cluster state with the manifests stored in Git. The DataScienceCluster controller manages the operator components but does not enforce external repository states.
More Red Hat Certified Specialist in OpenShift AI (EX267) drills and other practice exams are on @CertPunch. New rounds drop every few days at certpunch.com.