What should I focus on for a Kubernetes interview?
Which topics should I start with?
Start with basics and architecture. This topic appeared in 25 of 94 interviews, more than any other topic. Deployment, Helm and GitOps appeared in 23 of 94 interviews, as did troubleshooting and monitoring.
Use these topics as a starting checklist. Can you explain how cluster components work together? Can you describe how you would release an application, roll it back, and investigate a pod failure? These checks cover more than a definition of Kubernetes.
Can I explain Kubernetes and application configuration clearly?
“What is Kubernetes, and what is it used for?” was the most asked question, followed by the question about ConfigMaps and Secrets. Reviews of real answers found gaps in both topics.
Some answers did not clearly distinguish the control plane, worker nodes, pods, and containers. Check whether you can name these parts and explain how they fit together, rather than just list them.
Other answers did not clearly explain when to use ConfigMaps versus Secrets. Can you make this distinction?
- ConfigMaps: non-sensitive configuration.
- Secrets: sensitive values such as passwords, tokens, and API keys.
Your answer should cover when to use each, not just what each is called.
Can I talk through a pod failure and a release?
The CrashLoopBackOff question asked how you would investigate even when CPU and memory were available. Does your answer address that detail, or does it only discuss available resources?
Use these questions to check troubleshooting and delivery:
- How would you use logs and metrics to investigate a pod failure?
- How does Kubernetes handle rolling updates and rollbacks?
- What is a Helm chart?
- How would you move a release across environments?
Keep the failure investigation and the release process clear in your answer. Both were topics in these interviews.
Which other questions should I check?
Platform choices, networking, security, and scaling also appeared. Check whether you can answer these specific questions:
- How would you choose between EC2, ECS, and EKS?
- How do ClusterIP, NodePort, LoadBalancer, Ingress, and OpenShift Route differ, and when would you use each?
- What is the role of a Service?
- How do you use resource requests and limits?
- How would you use RBAC to control access?
For AI or data work, add workload-specific checks: how would you serve a model, allocate GPUs, or tune Spark on Kubernetes?
Who gets asked Kubernetes
The 20 most asked Kubernetes questions
Counted by interview, not by mention: a question asked twice in one conversation counts once.
What is Kubernetes, and what is it used for?
Asked in 9 of 94 interviews · last asked Oct 22 · Concept · Junior
Show answer
Kubernetes is an open-source platform for managing containerized applications. It automates deployment, scaling, and recovery when workloads fail. The control plane manages the cluster and schedules Pods onto worker nodes. Each Pod holds one or more containers that run together.
- Kubernetes works to keep running workloads matched to the desired state you define.
- It supports service discovery, load balancing, and rolling updates.
- It can restart failed containers and replace failed Pods.
- It manages containers but does not build container images.
What the interviewer checks: The interviewer is checking whether you understand Kubernetes, its main parts, and the problems it solves.
Common mistake (in 3 of 7 reviewed answers): Many answers forget that the control plane manages worker nodes, which run Pods containing containers.
Other ways it was asked
- “What is Kubernetes?”
- “What is Kubernetes, and have you worked with it?”
- “What is Kubernetes, and why is it used?”
What are ConfigMaps and Secrets, how do they differ, and when should you use each?
Asked in 7 of 94 interviews · last asked Oct 7 · Concept · Junior
Show answer
ConfigMaps store non-sensitive settings, such as service URLs and feature flags. Secrets store sensitive values, such as passwords, tokens, and API keys. Pods can use either as environment variables or mounted files. Secret values are base64-encoded, not encrypted by that encoding; use access controls and enable encryption at rest.
- Use ConfigMaps for ordinary settings and Secrets for credentials.
- Both keep configuration separate from application code and container images.
- Secrets are not encrypted in etcd by default; restrict access with RBAC and configure encryption at rest.
What the interviewer checks: The interviewer checks whether you can separate application settings from credentials and understand Secret security limits.
Common mistake (in 3 of 5 reviewed answers): Many answers forget that ConfigMaps are for non-sensitive settings, while passwords, tokens, and API keys belong in Secrets.
Other ways it was asked
- “What is a ConfigMap in Kubernetes?”
- “Explain what a ConfigMap is in Kubernetes.”
- “What is a Kubernetes ConfigMap?”
What are the Kubernetes control-plane and worker-node components, and how do they work together?
Asked in 5 of 94 interviews · last asked Sep 21 · Concept · Mid-level
Show answer
The control plane manages the cluster; worker nodes run Pods. The control plane includes kube-apiserver, etcd, kube-scheduler, and kube-controller-manager. Workers use kubelet, a container runtime, and usually kube-proxy. The API server accepts changes, etcd stores cluster state, the scheduler assigns Pods, controllers reconcile desired state, and kubelets run assigned Pods through the runtime.
- Components communicate through the API server; only the API server directly accesses etcd.
- Kubelet manages Pods on its node and reports their status to the API server.
- Kube-proxy implements Service networking rules; some network plugins replace it.
- An optional cloud-controller-manager connects Kubernetes to cloud services, such as load balancers.
What the interviewer checks: The interviewer checks whether you understand each component’s role and how they work together to run workloads.
Other ways it was asked
- “What are the Kubernetes control-plane components?”
- “What is the Kubernetes control plane, and what are its components?”
- “What is the Kubernetes control plane?”
How would you troubleshoot a pod in CrashLoopBackOff, even when CPU and memory are available?
Asked in 5 of 94 interviews · last asked Oct 7 · Scenario · Mid-level
Show answer
I would check the container’s previous logs, Pod events, and last termination reason and exit code. Then I would check its startup command, configuration, Secrets, dependencies, and health probes. Free CPU and memory do not rule out application errors, probe failures, or a container hitting its own memory limit.
- CrashLoopBackOff means Kubernetes is delaying repeated container restarts; it is not the root cause.
- Use previous logs because the current container may have just restarted.
- Failed startup or liveness probes can trigger restarts; failed readiness probes do not.
- Check for OOMKilled even when the node has free memory.
Codekubectl logs POD -c CONTAINER --previous kubectl describe pod POD kubectl get pod POD -o yamlWhat the interviewer checks: The interviewer is checking whether you use evidence to find the restart cause instead of assuming resource shortages.
Other ways it was asked
- “How would you troubleshoot a pod that is in CrashLoopBackOff?”
- “A Kubernetes pod is in CrashLoopBackOff. How would you diagnose the problem?”
- “Suppose an application is crashing randomly in Kubernetes. How would you troubleshoot it?”
How would you choose between EC2, ECS, and EKS for deploying an application?
Asked in 5 of 94 interviews · last asked Oct 14 · Concept · Mid-level
Show answer
I would choose EC2 for direct control of virtual machines, ECS for simpler AWS-managed container deployment, and EKS when Kubernetes is needed. ECS fits teams focused on AWS with less operational work. EKS fits teams using Kubernetes tools or seeking portability, but adds complexity. I would weigh team skills, workload needs, and total cost.
- EC2 provides virtual machines; ECS and EKS manage containers and can use EC2 for compute.
- ECS and EKS can also use Fargate for supported workloads, avoiding direct server management.
- EKS manages the Kubernetes control plane; the team still manages workloads and cluster configuration.
- Choose EKS for Kubernetes compatibility, not simply because the application uses containers.
What the interviewer checks: The interviewer checks whether you can match infrastructure choices to application needs, team skills, and operational cost.
Other ways it was asked
- “What is the difference between EC2, ECS, and EKS, and how would you use them?”
- “What is the difference between ECS and EKS, and in which situations would you choose one over the other?”
- “In AWS, how would you choose among a virtual machine, Kubernetes, or a containerized solution for deployment?”
How do ClusterIP, NodePort, LoadBalancer, Ingress, and OpenShift Route differ, and when would you use each?
Asked in 4 of 94 interviews · last asked Sep 29 · Concept · Mid-level
Show answer
ClusterIP exposes a Service inside the cluster for internal traffic. NodePort exposes it on node ports for direct access; LoadBalancer requests a load balancer when the environment supports it. Ingress routes HTTP/HTTPS traffic to Services by hostname or path and needs a controller. OpenShift Route provides similar web routing through an OpenShift router.
- ClusterIP, NodePort, and LoadBalancer are Service types; Ingress and Route are separate routing resources.
- Use NodePort for simple external access or your own load balancer; node addresses and ports must be reachable.
- Use LoadBalancer for a dedicated entry point; public or private access depends on provider configuration.
- Use Ingress or OpenShift Route to share web entry points across applications and manage TLS.
What the interviewer checks: The interviewer checks whether you can choose internal access, direct external access, or web routing for an application.
Other ways it was asked
- “What is the difference between ClusterIP and NodePort in Kubernetes, and why would you choose NodePort over ClusterIP?”
- “What is the difference between ClusterIP, NodePort, LoadBalancer, Route, and Ingress?”
- “Why would you choose a ClusterIP service over the other Kubernetes service types?”
What is the role of a Service in Kubernetes?
Asked in 4 of 94 interviews · last asked Jul 27 · Concept · Junior
Show answer
A Kubernetes Service gives a group of Pods a stable network address and DNS name. It sends traffic to the Pods even when they are replaced and their IP addresses change. By default, a Service is reachable only inside the cluster, but other Service types can expose it outside.
- A Service usually selects Pods using labels.
- It distributes traffic across ready Pod endpoints.
- ClusterIP provides internal access; NodePort and LoadBalancer can provide external access.
What the interviewer checks: The interviewer is checking whether you understand how Services provide stable access to changing Pods.
Other ways it was asked
- “What is the role of a Kubernetes service?”
- “What is a Service used for in Kubernetes?”
- “What is a Service in Kubernetes, and why is it needed?”
How does Kubernetes handle rolling updates and rollbacks?
Asked in 4 of 94 interviews · last asked Sep 4 · Concept · Mid-level
Show answer
For Deployments using `RollingUpdate`, Kubernetes rolls out a changed Pod template by gradually replacing old Pods with new ones. It uses readiness checks and update limits to control availability. A rollback restores a saved Pod template revision, usually with `kubectl rollout undo`. A failed rollout does not automatically trigger a rollback.
- A Deployment creates or reuses a ReplicaSet when its Pod template changes.
- `maxSurge` limits extra Pods; `maxUnavailable` limits unavailable Pods during the update.
- Readiness probes help prevent new Pods from receiving traffic before they are ready.
- Rollback requires a retained revision and does not undo database or external configuration changes.
What the interviewer checks: The interviewer checks whether you understand gradual Pod replacement, availability controls, and rollback limits.
Other ways it was asked
- “How do rolling updates work in Kubernetes?”
- “How does a rolling update work in Kubernetes?”
What is a Helm chart?
Asked in 4 of 94 interviews · last asked Sep 17 · Concept · Junior
Show answer
A Helm chart is a package of files used to define and install an application in Kubernetes. Helm is the package manager that uses these charts. A chart contains chart details, templates for Kubernetes resources, and default configuration values. Helm combines the templates with your values to install or update a release.
- Chart.yaml contains chart details, such as its name and version.
- templates/ holds Kubernetes resource templates; values.yaml provides default values.
- Charts can declare dependencies on other charts.
- A release is an installed instance of a chart; Helm supports upgrades and rollbacks.
What the interviewer checks: The interviewer checks whether you understand Helm, chart contents, and how charts relate to releases.
Other ways it was asked
- “What goes inside a Helm chart?”
- “What does a Helm chart contain?”
- “What is Helm?”
What is a deployment in Kubernetes?
Asked in 3 of 94 interviews · last asked Sep 4 · Concept · Junior
Show answer
A Deployment is a Kubernetes object that manages a set of Pods through ReplicaSets. It keeps the desired number of Pods running and supports controlled updates and rollbacks. You create one to run and update an application, usually one that does not need a stable identity or storage for each Pod.
- Defines the desired Pod template and replica count.
- Manages ReplicaSets, which create and replace Pods.
- Supports rolling updates and rollback to an earlier Pod template.
- Commonly used for stateless applications.
What the interviewer checks: The interviewer checks whether you understand how Deployments keep applications running and manage updates.
Other ways it was asked
- “What is a Deployment in Kubernetes, and what is its purpose?”
- “What is a Kubernetes Deployment, what is it used for, and why would you create one?”
How does the Kubernetes Horizontal Pod Autoscaler work when it is based on metrics?
Asked in 3 of 94 interviews · last asked Oct 7 · Concept · Mid-level
Show answer
The Horizontal Pod Autoscaler (HPA) changes a workload’s replica count based on observed metrics. It periodically compares current metrics with target values and calculates how many replicas are needed. It supports CPU, memory, custom metrics, and external metrics. Scaling stays within configured minimum and maximum replica counts.
- CPU utilization percentages are measured against container CPU requests, not CPU limits.
- Resource metrics usually come from Metrics Server; custom and external metrics need adapters.
- With multiple metrics, HPA uses the largest replica count recommended by any metric.
- Stabilization windows and scaling policies help prevent rapid changes in replica count.
What the interviewer checks: The interviewer checks whether you understand metric sources, target comparison, and controls that keep scaling stable.
Other ways it was asked
- “What is the Kubernetes Horizontal Pod Autoscaler, and how does it scale workloads?”
- “How does the Kubernetes Horizontal Pod Autoscaler (HPA) work, and which metrics can it use?”
How do you configure Kubernetes autoscaling, including metrics, utilization targets, and minimum and maximum replicas?
Asked in 3 of 94 interviews · last asked Oct 7 · Tooling · Mid-level
Show answer
I configure a Horizontal Pod Autoscaler (HPA) with metrics, target values, and minimum and maximum replicas. For CPU or memory utilization, I set container resource requests and make metrics available through Metrics Server. I choose targets and replica limits based on load tests, availability needs, and available capacity.
- CPU utilization is measured against CPU requests, not CPU limits.
- Custom metrics, such as queue length, need a metrics adapter.
- Minimum replicas protect baseline availability; maximum replicas cap scaling but do not guarantee capacity.
- HPA scales Pod replicas; node autoscaling is configured separately.
Code# Requires deployment web, CPU requests, and working resource metrics. # Example: target 70% average CPU utilization, with 2–10 replicas. kubectl autoscale deployment web --cpu-percent=70 --min=2 --max=10 kubectl get hpaWhat the interviewer checks: The interviewer checks whether you can configure HPA and explain its metric requirements, utilization targets, and scaling limits.
Other ways it was asked
- “How do you set up autoscaling in Kubernetes?”
- “How do you set up autoscaling in Kubernetes, and what metrics do you use?”
- “How do you configure scaling rules for workloads in Kubernetes?”
Can you explain the different deployment strategies available in Kubernetes?
Asked in 3 of 94 interviews · last asked Jul 16 · Concept · Mid-level
Show answer
Kubernetes Deployments support two built-in strategies: RollingUpdate and Recreate. RollingUpdate gradually replaces old Pods, while Recreate stops all old Pods before starting new ones. Blue-green and canary are rollout patterns built using separate workloads and traffic routing. Blue-green switches traffic between versions; canary sends a small share to the new version first.
- RollingUpdate is the default; maxSurge and maxUnavailable control extra Pods and allowed unavailability.
- Readiness probes help keep traffic away from Pods that are not ready.
- Recreate causes downtime but avoids running old and new Deployment Pods together.
- Blue-green enables quick traffic rollback but needs extra capacity; canary limits exposure but requires monitoring.
What the interviewer checks: The interviewer checks whether you understand rollout trade-offs and distinguish built-in Deployment strategies from broader release patterns.
Other ways it was asked
- “What types of deployments are available in Kubernetes?”
- “What deployment strategies are available in Kubernetes?”
How do you handle your Kubernetes cluster security?
Asked in 3 of 94 interviews · last asked Sep 4 · Scenario · Senior
Show answer
I secure a Kubernetes cluster in layers. I use trusted identity providers, least-privilege RBAC, and restricted access to the API server. I enforce Pod Security Standards, limit pod traffic with NetworkPolicies, and protect secrets. I also patch nodes and Kubernetes, scan container images, and monitor audit logs.
- Use separate service accounts and avoid giving users or workloads cluster-admin access.
- NetworkPolicies need a supporting network plugin; start with default-deny rules and allow required traffic.
- Restrict privileged containers and require non-root users where possible.
- Enable encryption at rest for Secrets; base64 encoding is not encryption.
What the interviewer checks: The interviewer checks whether you cover access control, workload isolation, data protection, and ongoing security maintenance.
Other ways it was asked
- “How do you secure a Kubernetes cluster?”
- “Can you explain how you manage access to your Kubernetes cluster?”
How does Kubernetes RBAC work, and how do Roles, ClusterRoles, and bindings differ?
Asked in 3 of 94 interviews · last asked Oct 7 · Concept · Mid-level
Show answer
Kubernetes RBAC controls who can perform which actions on API resources. A Role defines permissions in one namespace; a ClusterRole can define permissions for namespaced or cluster-wide resources. Bindings grant these permissions to users, groups, or service accounts. A RoleBinding grants access within one namespace; a ClusterRoleBinding grants access across the cluster.
- Rules specify API groups, resources, and verbs, such as get, list, or create.
- A RoleBinding references a Role in its own namespace or a ClusterRole, but grants only namespaced access.
- A ClusterRoleBinding references only a ClusterRole.
- RBAC permissions add together. There are no deny rules; without a matching grant, RBAC denies access.
What the interviewer checks: The interviewer checks whether you understand permission rules, binding subjects, and namespace versus cluster scope.
Other ways it was asked
- “How does Kubernetes RBAC work?”
- “What is the difference between a Kubernetes Role and a ClusterRole?”
- “What is Kubernetes RBAC, and what are Roles, ClusterRoles, and RoleBindings?”
Why might a Kubernetes pod remain Pending, and how would you troubleshoot it?
Asked in 3 of 94 interviews · last asked Sep 4 · Scenario · Mid-level
Show answer
A pod stays Pending when Kubernetes cannot schedule it or has not finished setting up its containers. I check its assigned node, then read `kubectl describe pod` events. I look for resource shortages, scheduling restrictions, unbound volumes, or image and network errors. I fix the reported cause and check the pod again.
- Scheduling uses resource requests, not current CPU or memory usage.
- Check node readiness, taints, tolerations, node selectors, and affinity rules.
- Check whether persistent volume claims are waiting for suitable storage.
- If a node is assigned, inspect container status and events for image pull or network setup failures.
Codekubectl get pod my-pod -n my-namespace -o wide kubectl describe pod my-pod -n my-namespace kubectl get pvc -n my-namespace kubectl get nodesWhat the interviewer checks: The interviewer checks whether you use status and events to distinguish scheduling failures from container setup problems.
Other ways it was asked
- “A pod is stuck in a pending state. How do you debug this scenario?”
- “How do you debug a Kubernetes pod that is stuck in a pending state?”
- “How do you debug a pod stuck in a pending state?”
What is the difference between a Docker image and a container, and between Docker and Kubernetes?
Asked in 2 of 94 interviews · last asked Sep 30 · Concept · Junior
Show answer
A Docker image is a read-only template; a container is an instance of that image. Docker builds images and runs containers. Kubernetes manages container workloads across machines, handling scheduling, scaling, and recovery. Kubernetes deploys Pods, not individual containers: a Pod holds one or more containers that share networking and can share storage.
- A container can be running or stopped; the image itself does not run.
- Containers have writable layers; changes there do not change the image.
- Containers in one Pod run on the same node and can communicate through localhost.
- Kubernetes does not require Docker Engine; it uses compatible container runtimes, such as containerd.
What the interviewer checks: The interviewer checks whether you distinguish images, containers, Pods, and the roles of Docker and Kubernetes.
Other ways it was asked
- “What is the difference between a Docker container and a Kubernetes pod?”
What is a Kubernetes node?
Asked in 2 of 94 interviews · last asked Sep 4 · Concept · Junior
Show answer
A Kubernetes node is a physical or virtual machine in a Kubernetes cluster. Worker nodes run application Pods, while control plane nodes manage the cluster. Each worker node uses a kubelet and a container runtime to run and manage containers.
- A Pod runs on one node; a node can run many Pods.
- The kubelet makes sure assigned Pods are running as expected.
- The scheduler selects a node for each Pod based on available resources and scheduling rules.
What the interviewer checks: The interviewer is checking whether you understand what nodes do and how they relate to Pods and the control plane.
Other ways it was asked
- “What is a node in Kubernetes?”
What is the purpose of namespaces in Kubernetes?
Asked in 2 of 94 interviews · last asked Oct 7 · Concept · Junior
Show answer
Namespaces divide a Kubernetes cluster into named groups of resources. They help teams or projects share a cluster while managing access and resource limits separately. Namespaces alone do not isolate network traffic.
- Resource names must be unique within a namespace for each resource type.
- RBAC controls access, and ResourceQuota limits resource use within a namespace.
- Some resources, such as nodes and persistent volumes, are cluster-wide.
- NetworkPolicies can restrict traffic if the network plugin supports them.
What the interviewer checks: The interviewer is checking whether you understand resource grouping, namespace scope, and the limits of namespace isolation.
Other ways it was asked
- “What is a Kubernetes namespace?”
What is the difference between a Deployment and a StatefulSet in Kubernetes?
Asked in 2 of 94 interviews · last asked Oct 7 · Concept · Junior
Show answer
A Deployment manages interchangeable Pods, usually for stateless applications. A StatefulSet gives each Pod a stable name and network identity, and can give it its own persistent storage. Deployments support rolling updates with extra Pods, while StatefulSets normally create, update, and remove Pods in order.
- Deployment Pods are interchangeable; StatefulSet Pods keep their identity when replaced.
- StatefulSet volume claim templates give each Pod separate persistent storage that can survive Pod replacement.
- StatefulSets use a headless Service for stable Pod network names, not stable IP addresses.
- Deployments can also use persistent volumes; StatefulSets do not automatically replicate data.
What the interviewer checks: The interviewer checks whether you can choose between interchangeable replicas and Pods needing stable identity or storage.
Other ways it was asked
- “What is the difference between a Kubernetes Deployment and a StatefulSet?”
Kubernetes basics and architecture interview questions
Start with questions about Kubernetes uses, containers, cluster components, namespaces, kubectl, controllers, CRDs, and operators. 23 questions.
What are the benefits and weaknesses of Kubernetes?
Asked once · last asked May 7 · Concept · Mid-level
Other ways it was asked
- “Based on your experience, what are the benefits and weaknesses of Kubernetes?”
- “What are the benefits and drawbacks of Kubernetes?”
When would you recommend Kubernetes over alternatives, considering scalability, cost, and time to launch?
Asked once · last asked Feb 11 · Scenario · Mid-level
Other ways it was asked
- “Why would you choose Kubernetes, and what alternatives would you consider?”
- “What requirements would lead you to recommend Kubernetes, considering scalability, cost, and time to go live?”
If you are not using a microservices architecture, would you still use Kubernetes?
Asked once · last asked Feb 11 · Scenario · Mid-level
Can you talk about how Docker and Kubernetes together contribute to the scalability and resilience of microservices architectures?
Asked once · last asked Mar 25 · Concept · Mid-level
What types of applications are commonly run on Kubernetes, such as internal applications, databases, or web applications?
Asked once · last asked May 7 · Concept · Mid-level
What is the role of the first node in a Kubernetes cluster, and how does it differ from the other nodes?
Asked once · last asked May 12 · Concept · Mid-level
How would you use kubectl get to list Kubernetes resources?
Asked once · last asked May 18 · Tooling · Junior
What problems does Docker solve, and what is Kubernetes used for?
Asked once · last asked May 22 · Concept · Mid-level
What is etcd, and what role does it play in a Kubernetes cluster?
Asked once · last asked Jul 16 · Concept · Mid-level
What is the difference between a pod, a Deployment, and a Service in Kubernetes?
Asked once · last asked Oct 7 · Concept · Mid-level
What are Custom Resource Definitions (CRDs) in Kubernetes?
Asked once · last asked Sep 4 · Concept · Mid-level
Other ways it was asked
- “Why would you use a Custom Resource Definition (CRD) in Kubernetes?”
What is a Kubernetes operator?
Asked once · last asked Sep 4 · Concept · Mid-level
Other ways it was asked
- “Why would you use a Kubernetes operator?”
Show 6 more
Why do we use Kubernetes with Docker?
Asked once · last asked Mar 3 · Concept · Junior
Which command-line tool is used to work with Kubernetes?
Asked once · last asked Jul 27 · Concept · Junior
What is a custom controller?
Asked once · last asked Sep 4 · Concept · Junior
Other ways it was asked
- “How would you build a custom controller?”
What are the main components of a Kubernetes worker node?
Asked once · last asked Sep 21 · Concept · Junior
What kubectl command lists Pods in Kubernetes?
Asked once · last asked Oct 7 · Tooling · Junior
What is the main difference between Kubernetes and containers?
Asked once · last asked Oct 22 · Concept · Junior
Kubernetes pods, workloads and storage interview questions
Explore how pods, workload types, sidecars, application configuration, and persistent storage fit into running applications. 21 questions.
What is a DaemonSet?
Asked in 2 of 94 interviews · last asked Sep 4 · Concept · Junior
Other ways it was asked
- “What is a DaemonSet in Kubernetes?”
What is a Kubernetes pod, and why use one instead of running a container directly?
Asked in 2 of 94 interviews · last asked Oct 7 · Concept · Junior
Other ways it was asked
- “What are pods in Kubernetes?”
How would you run dynamic background jobs as Kubernetes Jobs or Rails processes, consume work from a queue, and handle long-running tasks?
Asked once · last asked Jul 24 · Design · Senior
Other ways it was asked
- “How do dynamic background jobs run as Kubernetes Jobs or Rails processes, pick up work from a queue, and handle long-running tasks?”
How would you distinguish stateful from stateless applications and choose a Kubernetes workload setup, such as a StatefulSet, a stateless Deployment, or a DaemonSet, for a database like SQL Server versus a JavaScript front end?
Asked once · last asked Feb 11 · Scenario · Senior
How do you use an OpenShift ConfigMap to configure environment variables for an application?
Asked once · last asked Apr 28 · Tooling · Mid-level
How would you deploy a WAR file to one or more Kubernetes pods?
Asked once · last asked May 12 · Scenario · Mid-level
How do you orchestrate multiple containers in a Kubernetes cluster?
Asked once · last asked May 14 · Tooling · Mid-level
When moving an application from a computer instance to Kubernetes, what should its container image contain?
Asked once · last asked May 14 · Concept · Mid-level
How would you use Docker and Kubernetes to deploy an application across multiple servers?
Asked once · last asked Jun 15 · Tooling · Mid-level
How would you containerize and deploy a Java application to Kubernetes?
Asked once · last asked Aug 4 · Scenario · Mid-level
What are StatefulSets and DaemonSets in Kubernetes, and how do they differ?
Asked once · last asked Aug 4 · Concept · Mid-level
What is a sidecar container, and when should you use one?
Asked once · last asked Sep 4 · Concept · Mid-level
Show 6 more
Explain what a pod is.
Asked once · last asked Jul 25 · Concept · Junior
In Kubernetes, what does the “set” in DaemonSet mean, and how does it affect pod placement?
Asked once · last asked Aug 4 · Concept · Junior
What is a StatefulSet in Kubernetes?
Asked once · last asked Sep 4 · Concept · Mid-level
What are Persistent Volumes and Persistent Volume Claims, and how do they work together?
Asked once · last asked Sep 4 · Concept · Junior
Other ways it was asked
- “What is a Persistent Volume in Kubernetes?”
- “What are Persistent Volumes (PVs) and Persistent Volume Claims (PVCs) in Kubernetes?”
What are Kubernetes volumes used for?
Asked once · last asked Oct 7 · Concept · Junior
Which version of Spring Boot is most suitable for use with Kubernetes today?
Asked once · last asked Oct 18 · Concept · Mid-level
Kubernetes networking and Services interview questions
Networking questions cover pod communication, Services, ingress, load balancers, service meshes, DNS, and request routing. 22 questions.
What is a Kubernetes headless Service, and when would you use one?
Asked in 2 of 94 interviews · last asked Aug 7 · Concept · Mid-level
Other ways it was asked
- “What is a headless service in Kubernetes, and when would you use one?”
Walk me through a request from a user's device to an application pod in Kubernetes.
Asked in 2 of 94 interviews · last asked Sep 29 · Design · Mid-level
Other ways it was asked
- “Can you walk me through the workflow for a request to an API hosted on AKS, especially the components it passes through inside AKS before reaching the pod?”
- “What components does an API request pass through inside AKS before reaching a pod?”
- “Explain the Kubernetes request flow from a user's device to the application.”
Which networks and subnets must you plan for in Kubernetes, and what is each used for?
Asked in 2 of 94 interviews · last asked Sep 4 · Concept · Mid-level
Other ways it was asked
- “What other networks, such as subnets, must you plan for in a Kubernetes cluster?”
- “What networks are used in a Kubernetes cluster, and what is each one for?”
- “What other networks, besides the Service and Pod networks, are part of a basic Kubernetes architecture, and what are they used for?”
How do you expose a Kubernetes application to the Internet?
Asked in 2 of 94 interviews · last asked Sep 4 · Tooling · Mid-level
Other ways it was asked
- “How can you make a deployed Kubernetes application accessible over the Internet?”
- “How do you expose a Kubernetes application externally?”
How do you integrate an AWS Application Load Balancer with Kubernetes running on EKS?
Asked once · last asked Sep 29 · Scenario · Senior
Other ways it was asked
- “How do you connect an AWS Application Load Balancer to Kubernetes services in EKS?”
How does a Kubernetes Service route requests to healthy pods, and how are unhealthy pods excluded from service?
Asked once · last asked Apr 14 · Concept · Mid-level
What DNS configuration, including conditional forwarding, is needed to keep a private AKS cluster reachable if its IP changes after a restart?
Asked once · last asked Apr 14 · Scenario · Senior
What port range is available for Kubernetes NodePort services? Can NodePort use ports 0–62,000, or is there a limit?
Asked once · last asked May 12 · Concept · Mid-level
Your application is deployed in a Kubernetes cluster, and the database runs inside the cluster. How does communication between them work?
Asked once · last asked Aug 4 · Concept · Junior
How do two different containers running in a single pod, sharing the same IP address, communicate with each other?
Asked once · last asked Sep 4 · Concept · Mid-level
What is a service mesh, and why do we need it?
Asked once · last asked Sep 4 · Concept · Mid-level
What does kube-proxy do, how does it route Service traffic, and which backends can it use?
Asked once · last asked Sep 4 · Concept · Mid-level
Other ways it was asked
- “What network communication does kube-proxy provide in a Kubernetes cluster, and why is it needed?”
- “How does kube-proxy connect Kubernetes networks, and which backends can it use?”
Show 8 more
What is the Kubernetes Service network, and how does it work?
Asked once · last asked Sep 4 · Concept · Mid-level
How would you choose between a custom ingress controller and Kubernetes ingress controllers such as NGINX or Ambassador?
Asked once · last asked Sep 22 · Design · Senior
What targets need to be registered in an AWS load balancer target group for EKS?
Asked once · last asked Sep 29 · Concept · Senior
What component does Kubernetes need to provide networking?
Asked once · last asked May 12 · Concept · Mid-level
Other ways it was asked
- “What does Kubernetes need to provide networking to pods?”
Does a Kubernetes Service restart or terminate a pod's process?
Asked once · last asked Apr 14 · Concept · Junior
What type of network do you configure in Kubernetes?
Asked once · last asked May 12 · Concept · Mid-level
Which service ports need to be registered in the target group?
Asked once · last asked Sep 29 · Concept · Mid-level
Which pods should be registered in the target group?
Asked once · last asked Sep 29 · Concept · Mid-level
Kubernetes scheduling and autoscaling interview questions
Check your understanding of pod placement, resource requests and limits, memory behavior, autoscaling metrics, replica counts, and capacity planning. 15 questions.
How would you scale a Kubernetes application to handle increased demand?
Asked in 2 of 94 interviews · last asked Sep 4 · Tooling · Mid-level
Other ways it was asked
- “How would you scale a Kubernetes pod to handle heavy demand?”
- “How do you scale a Kubernetes deployment?”
If the system experiences heavy load every day from 3 to 4 p.m. and response times exceed the SLA, how should you handle it using AWS or Kubernetes?
Asked once · last asked Jan 30 · Scenario · Senior
Does a Kubernetes Deployment automatically scale its replicas when a pod becomes unavailable?
Asked once · last asked Apr 14 · Concept · Mid-level
Where and how do you configure CPU and memory resources for an AKS workload?
Asked once · last asked Apr 14 · Tooling · Mid-level
Other ways it was asked
- “How do you configure resource allocations for an API workload in AKS, and where do you specify them?”
- “Where in an AKS workload configuration do you specify CPU and memory resources?”
What are Kubernetes resource requests and limits, and how do they affect pod scheduling and resource use?
Asked once · last asked Apr 14 · Concept · Mid-level
Other ways it was asked
- “Where do you configure resource requests and limits for a Kubernetes pod?”
Given a 1 GB memory request, a 2 GB limit, and an HPA target of 70%, what memory usage would trigger scaling?
Asked once · last asked Apr 14 · Scenario · Mid-level
A node has 15 GB of available memory, and each pod requests 1 GB and has a 2 GB limit. How many such pods can Kubernetes schedule on the node?
Asked once · last asked Apr 14 · Scenario · Mid-level
How does a Kubernetes pod behave when it has a 2 GB memory limit?
Asked once · last asked Apr 14 · Scenario · Mid-level
If multiple Kubernetes pods approach 1 GB of memory usage, what happens under memory pressure, and can one pod's limit or usage cause other pods to be killed?
Asked once · last asked Apr 14 · Scenario · Senior
How does Kubernetes decide which node to run a Pod on?
Asked once · last asked May 7 · Concept · Mid-level
How would you design a data-processing application to scale horizontally as pods are added or removed, so performance scales with the number of pods?
Asked once · last asked May 19 · Design · Senior
In a Kubernetes cluster with nodes of different sizes, how would you ensure your application runs on the appropriate nodes?
Asked once · last asked Aug 4 · Scenario · Mid-level
Show 1 more
How would you prevent or limit runaway processes in Kubernetes?
Asked once · last asked Oct 7 · Scenario · Senior
Kubernetes deployment, Helm and GitOps interview questions
Deployment questions cover manifests, Helm charts, overlays, CI/CD, Argo CD, rollouts, rollbacks, and moving releases across environments. 27 questions.
How would you deploy Kubernetes application updates without downtime?
Asked in 2 of 94 interviews · last asked Sep 4 · Scenario · Senior
Other ways it was asked
- “How can you deploy your application on Kubernetes with zero downtime?”
- “A Kubernetes deployment update has caused downtime. How would you prevent this?”
How is a Helm chart organized, and what do its files and directories contain?
Asked in 2 of 94 interviews · last asked Sep 1 · Tooling · Mid-level
Other ways it was asked
- “How do you organize and manage the files used to deploy applications with Helm charts?”
- “What is the step-by-step deployment flow for managing Helm chart files and values?”
- “What files and resources are typically included in a Helm chart?”
How would you design and deploy two microservices from separate Azure Repos repositories to AKS, including the architecture, deployment steps, and tools?
Asked once · last asked Jan 31 · Design · Senior
When would you use GitOps versus deploying Helm charts from a CI/CD pipeline?
Asked once · last asked Mar 19 · Concept · Senior
What command do you use to restart a deployment in OpenShift?
Asked once · last asked Apr 28 · Tooling · Mid-level
How would you package Helm charts for platform services?
Asked once · last asked Jul 16 · Tooling · Senior
What do maxSurge and maxUnavailable control during a Kubernetes rolling update?
Asked once · last asked Aug 5 · Concept · Mid-level
How do you deploy applications to IBM Kubernetes Service (IKS) using Helm?
Asked once · last asked Aug 7 · Tooling · Mid-level
How can you automate application deployments to AKS?
Asked once · last asked Aug 7 · Scenario · Mid-level
How do you build Docker images, write Kubernetes manifests, and create deployment guides for a microservice?
Asked once · last asked Aug 31 · Tooling · Mid-level
How do you implement GitOps to synchronize Kubernetes manifests in Git with a cluster?
Asked once · last asked Sep 4 · Tooling · Senior
Other ways it was asked
- “How do you implement GitOps in Kubernetes?”
- “How do you use a GitOps tool to store Kubernetes manifests in Git and synchronize them with a cluster?”
How would you design the deployment of artifacts into an AWS EKS Kubernetes cluster?
Asked once · last asked Sep 17 · Design · Senior
Show 12 more
What is Argo CD, and how does it differ from other continuous delivery tools?
Asked once · last asked Sep 25 · Concept · Mid-level
How would you manage service rollouts and updates on Amazon EKS, and mitigate cold-start problems?
Asked once · last asked Sep 29 · Scenario · Senior
What is the difference between a Kubernetes deployment strategy and a PodDisruptionBudget (PDB)?
Asked once · last asked Sep 29 · Concept · Senior
How would you deploy and operate an application end to end, including the build and image registry, Argo CD synchronization, rollout strategy, and promotion across development, staging, and production?
Asked once · last asked Oct 1 · Tooling · Senior
How do you organize Kubernetes configuration into a base and environment-specific overlays?
Asked once · last asked Oct 1 · Tooling · Mid-level
How does Argo CD determine which Kubernetes configuration file supplies the image reference for a deployment?
Asked once · last asked Oct 1 · Concept · Mid-level
How do you handle failures and rollbacks when a CI workflow updates a Kubernetes overlay and Argo CD syncs it?
Asked once · last asked Oct 1 · Scenario · Senior
How do you ensure changes in GKE Autopilot clusters do not disrupt your CI/CD pipeline?
Asked once · last asked Oct 3 · Scenario · Senior
How do you use Argo CD and Argo Rollouts to manage production deployments?
Asked once · last asked Oct 20 · Tooling · Senior
When deploying on Kubernetes, do you use Helm or customize more, and why?
Asked once · last asked Oct 20 · Tooling · Mid-level
Is a Helm chart best understood as a package or as a configuration?
Asked once · last asked Sep 1 · Concept · Junior
How does Argo CD differ from a CI runner?
Asked once · last asked Sep 25 · Concept · Mid-level
Kubernetes security interview questions
Review questions about RBAC, Secrets, certificates, external vaults, network isolation, admission controls, runtime protection, and security assessments. 15 questions.
How do you obtain and configure a certificate in Kubernetes for use across multiple ports?
Asked in 2 of 94 interviews · last asked Jul 25 · Tooling · Mid-level
Other ways it was asked
- “How do you configure certificates in a Kubernetes cluster?”
How do you secure Kubernetes Secrets?
Asked in 2 of 94 interviews · last asked Sep 22 · Tooling · Mid-level
Other ways it was asked
- “How do you get the secrets from Key Vault to Kubernetes?”
How can you use Falco and system calls to protect the runtime environment in Kubernetes?
Asked once · last asked Sep 1 · Tooling · Senior
Other ways it was asked
- “How can you use Falco and system calls to protect the runtime environment in Kubernetes, and write rules to detect Bash sessions or binary overwrites?”
How do you retrieve an Azure Key Vault certificate into an AKS Secret, and which driver is required?
Asked once · last asked Apr 14 · Tooling · Senior
Other ways it was asked
- “How would you integrate an SSL certificate from Azure Key Vault with AKS so it is available as a Kubernetes secret for ingress, and what driver is required?”
- “Which component or driver is needed to retrieve a certificate from Azure Key Vault into AKS?”
After gaining control of a Kubernetes pod during a penetration test, what post-exploitation steps would you consider?
Asked once · last asked May 14 · Scenario · Senior
What techniques could an attacker use to gain control of a Kubernetes pod and escape its container?
Asked once · last asked May 14 · Scenario · Senior
How do you secure traffic between OpenShift and external object storage?
Asked once · last asked Jul 16 · Scenario · Senior
How would you manage dynamic nodes, such as Kubernetes or Docker endpoints, through CyberArk?
Asked once · last asked Jul 22 · Scenario · Senior
What do you look for when assessing Kubernetes or containers, and how would you approach assessing an AI or LLM-based application?
Asked once · last asked Aug 21 · Scenario · Senior
How would Kubernetes workloads retrieve database connection strings from Azure Key Vault, and what role can managed identity play?
Asked once · last asked Sep 22 · Scenario · Senior
How would you design network isolation for Kubernetes clusters while allowing required access to other resources?
Asked once · last asked Sep 22 · Design · Senior
What is a network policy in Kubernetes?
Asked once · last asked Sep 4 · Concept · Junior
Other ways it was asked
- “What do Kubernetes network policies define, and how do they help secure a cluster?”
Show 1 more
What are Kubernetes admission controllers?
Asked once · last asked Sep 4 · Concept · Mid-level
Kubernetes troubleshooting and monitoring interview questions
Work through questions on pod failures, scheduling and connection problems, logs, metrics, health probes, self-healing, and disaster recovery. 31 questions.
How would you troubleshoot a Kubernetes Service that is not routing traffic to the correct pods?
Asked in 2 of 94 interviews · last asked Sep 4 · Scenario · Mid-level
Other ways it was asked
- “How do you debug a Kubernetes Service that is not reaching the correct pods?”
How would you troubleshoot an application on AKS when many of its pods are Pending despite available CPU and no clear issues in the events?
Asked once · last asked Aug 7 · Scenario · Mid-level
Other ways it was asked
- “What could cause application pods on AKS to remain Pending despite available CPU and no clear issues in the events?”
How would you select and integrate logging and monitoring tools with an AKS or Kubernetes cluster?
Asked once · last asked Jan 31 · Tooling · Mid-level
Other ways it was asked
- “What logging and monitoring tools have you used, and how do you integrate them with an AKS or Kubernetes cluster?”
How would you get custom metrics from different AKS namespaces into a Dynatrace dashboard, including deciding which metrics to define and transfer?
Asked once · last asked Jan 31 · Tooling · Senior
Other ways it was asked
- “How do you get custom metrics from different AKS namespaces into the Dynatrace dashboard? What metrics would you transfer, and have you defined any custom metrics?”
If you had many Kubernetes pods and wanted one pod to collect and consolidate logs from all the applications, what would you do?
Asked once · last asked Feb 11 · Scenario · Senior
How would you set up self-healing in Kubernetes so that requests are handled while a failing component recovers?
Asked once · last asked Feb 11 · Scenario · Senior
How do you deploy a Prometheus agent in Kubernetes?
Asked once · last asked Feb 23 · Tooling · Mid-level
Suppose a production Kubernetes cluster's control plane is down, the worker nodes are running, and the API server is unreachable. What would you do?
Asked once · last asked Mar 3 · Scenario · Senior
What is the difference between a liveness probe and a readiness probe in Kubernetes?
Asked once · last asked Mar 6 · Concept · Mid-level
What do the Kubernetes pod status reasons `OOMKilled` and `OutOfMemory` indicate?
Asked once · last asked Apr 14 · Concept · Mid-level
If a private AKS cluster is configured correctly but fails to start during creation, what network-level configuration might be needed?
Asked once · last asked Apr 14 · Scenario · Senior
How do you diagnose performance bottlenecks in a Kubernetes workload?
Asked once · last asked May 14 · Scenario · Mid-level
Show 17 more
How would you approach a database migration or troubleshoot a complex Kubernetes issue?
Asked once · last asked Jun 30 · Scenario · Senior
A deployment succeeds in CI/CD, but some components fail in OpenShift. What would you check?
Asked once · last asked Jul 16 · Scenario · Senior
How would you design disaster recovery for Kubernetes compute and object storage in a multi-tenant data platform?
Asked once · last asked Jul 16 · Design · Senior
What common Kubernetes issues would you troubleshoot, and how would you diagnose them?
Asked once · last asked Jul 24 · Behavioral · Mid-level
Other ways it was asked
- “What Kubernetes issue have you most commonly had to troubleshoot, and how did you diagnose it?”
If an error occurs in OpenShift, how would you debug and investigate it?
Asked once · last asked Jul 28 · Scenario · Mid-level
How does Kubernetes handle node failures?
Asked once · last asked Sep 4 · Concept · Mid-level
How do you troubleshoot a memory leak in a Kubernetes application?
Asked once · last asked Sep 4 · Scenario · Senior
Your Kubernetes cluster is running slow. How do you troubleshoot and optimize it?
Asked once · last asked Sep 4 · Scenario · Senior
How would you troubleshoot a production Kubernetes issue, and what responsibilities should you take during the incident?
Asked once · last asked Sep 10 · Scenario · Mid-level
How would you handle memory issues, violations, and exposed vulnerabilities in a Kubernetes cluster?
Asked once · last asked Sep 10 · Scenario · Senior
What would users experience if one of three production pods were deleted while handling traffic?
Asked once · last asked Sep 29 · Scenario · Mid-level
What would users experience if the only replica of an application were deleted during live traffic?
Asked once · last asked Sep 29 · Scenario · Mid-level
How would you investigate intermittent network connectivity between a Kubernetes pod and a database?
Asked once · last asked Sep 29 · Scenario · Senior
What happens when a Kubernetes pod keeps restarting?
Asked once · last asked Sep 30 · Scenario · Mid-level
After deploying an application to Kubernetes, what would you verify?
Asked once · last asked Aug 4 · Scenario · Junior
Does EKS provide an automatic recovery process?
Asked once · last asked Oct 1 · Concept · Mid-level
What happens if a pod managed by a Kubernetes Deployment is deleted?
Asked once · last asked Oct 7 · Concept · Junior
Kubernetes cluster management, EKS and AKS interview questions
Cluster management questions cover provisioning, managed and self-managed clusters, cloud choices, OpenShift, multiple clusters, and operational responsibilities. 20 questions.
In Amazon EKS, are node groups managed separately from the cluster, or does the cluster manage them?
Asked once · last asked Aug 5 · Concept · Mid-level
Why would an on-premises Kubernetes setup not be as scalable or resource-efficient?
Asked once · last asked Feb 11 · Concept · Senior
What are the differences between public and private AKS clusters?
Asked once · last asked Apr 14 · Concept · Mid-level
What operational challenges and trade-offs arise when moving from Kubernetes-based environments to bare-metal infrastructure?
Asked once · last asked Apr 22 · Design · Senior
How would you design and implement an end-to-end Google Cloud solution using a GKE cluster, and what components would it include?
Asked once · last asked Apr 24 · Design · Senior
What tools can you use to provision a Kubernetes node?
Asked once · last asked May 12 · Tooling · Mid-level
When provisioning a Kubernetes node, how do you install Linux packages, and when would you use the standard package manager versus another repository?
Asked once · last asked May 12 · Tooling · Mid-level
After provisioning the nodes, what else must you configure before a Kubernetes cluster is ready to use?
Asked once · last asked May 12 · Concept · Mid-level
How would you manage a Kubernetes setup, and when would you use a managed cluster versus building one from scratch?
Asked once · last asked May 14 · Concept · Mid-level
How would you decide whether to run Axway Secure Transport in an EKS cluster or directly on Linux?
Asked once · last asked Jun 18 · Scenario · Mid-level
Other ways it was asked
- “How would you decide whether to run Axway Secure Transport in a containerized environment or outside Kubernetes?”
How should responsibility for developing and managing a Kubernetes cluster be divided between an internal team and a vendor?
Asked once · last asked Jul 29 · Scenario · Senior
What is Auto Mode in Amazon EKS?
Asked once · last asked Aug 5 · Concept · Mid-level
Show 7 more
What prerequisites are needed to deploy an AKS cluster, and how does the handshake work?
Asked once · last asked Sep 3 · Concept · Senior
How would you rename an AKS cluster after it has been created?
Asked once · last asked Sep 3 · Scenario · Mid-level
How do you handle multi-cluster Kubernetes deployments?
Asked once · last asked Sep 4 · Design · Senior
How do you use Kubernetes in your day-to-day work?
Asked once · last asked Oct 3 · Tooling · Mid-level
What tasks are involved in operating Azure Kubernetes Service (AKS)?
Asked once · last asked May 7 · Tooling · Mid-level
Could you talk about some trade-offs with EKS?
Asked once · last asked Aug 13 · Concept · Senior
What is Azure Kubernetes Service?
Asked once · last asked Aug 21 · Concept · Junior
Kubernetes AI, GPU and data workload interview questions
For AI and data workloads, questions cover model serving, MLOps, testing, GPU allocation, Spark tuning, data pipelines, and specialized runtimes. 14 questions.
How do you deploy AI models on Kubernetes, and what other deployment tools would you consider?
Asked in 2 of 94 interviews · last asked Sep 24 · Tooling · Mid-level
Other ways it was asked
- “How do you deploy a model to Kubernetes?”
- “How do you use Kubernetes to deploy AI models, and what other tools or platforms do you use for different use cases?”
How would you trigger Spark on Kubernetes to process PostgreSQL staging tables populated by Flink without relying on scheduled orchestration?
Asked once · last asked May 18 · Design · Senior
Can machine learning models be deployed horizontally in Kubernetes, and how?
Asked once · last asked Jul 16 · Design · Mid-level
How do you tune Spark jobs running on Kubernetes or OpenShift?
Asked once · last asked Jul 16 · Tooling · Senior
If Spark executors or pods are not being scheduled, what would you check?
Asked once · last asked Jul 16 · Scenario · Senior
How do you manage Python and Conda environments in an OpenShift cluster, and how do you create a Conda environment there?
Asked once · last asked Jul 16 · Tooling · Senior
Would you deploy the pipeline components to a Kubernetes cluster and run them as long-lived services?
Asked once · last asked Jul 20 · Design · Senior
When deploying a FastAPI service to Kubernetes, what readiness and liveness probes and autoscaling signals would you choose to protect latency for an LLM/RAG endpoint?
Asked once · last asked Aug 24 · Design · Senior
How would you deploy a computer vision application with Docker and Kubernetes on AWS, and manage its MLOps pipeline?
Asked once · last asked Aug 27 · Design · Senior
What are the trade-offs between Kubernetes and managed endpoints for AI workloads?
Asked once · last asked Sep 24 · Concept · Senior
Other ways it was asked
- “What are the benefits of testing AI workloads on Kubernetes rather than using a managed endpoint?”
- “What are the trade-offs between a managed endpoint and Kubernetes for deploying AI workloads?”
When would you use GPU time slicing versus GPU passthrough for Kubernetes containers?
Asked once · last asked Oct 2 · Concept · Senior
What is the difference in accessing GPUs between Kubernetes and OpenStack?
Asked once · last asked Oct 2 · Concept · Senior
Show 2 more
What are the strengths and limitations of Red Hat OpenShift AI with NVIDIA?
Asked once · last asked Aug 20 · Concept · Mid-level
What does exploratory testing of AI inference workloads on Kubernetes involve?
Asked once · last asked Sep 24 · Tooling · Mid-level
Related interview questions
- DevOps Engineer interview questions
- Software Engineer interview questions
- Cloud Engineer interview questions
- Data Engineer interview questions
- Solutions Architect interview questions
- Data Scientist interview questions
- System Administrator interview questions
- AWS interview questions
- Azure interview questions
- Docker interview questions
- Terraform interview questions
- CI/CD interview questions
- Linux interview questions
- Active Directory and Windows Server interview questions
How this list is made
Every question here was asked by a real interviewer in a real job interview that a candidate ran MeetAssist in. After the call, the interviewer's questions are pulled from the transcript. Practice sessions, tests, exams and anything that was not a job interview are dropped.
Questions are rewritten so they make sense on their own: names of people, employers and clients are removed, and speech-to-text mistakes are fixed. Small talk, logistics and questions about one candidate's own résumé are left out. Rephrasings of the same question are merged, and the count is the number of different interviews it came up in.
The interviews come from candidates who use MeetAssist, so they lean towards software, data and AI roles: the shares describe these interviews, not every Kubernetes interview. Answers to the most asked questions are written and checked by an editor; a "common mistake" is shown only when reviews of real answers in these interviews show it, with the count.
94 interviews, 62 candidates, Jan 2026 to Oct 2026. Updated every month with new interviews.
Or add the Chrome extension