Interview questions

Cloud Computing Interview Questions

16 cloud computing interview questions on service models, shared responsibility, scaling, security, cost and resilience, with model-answer guidance.

Cloud Computing Interview Questions

What's here

16 Cloud Computing interview questions, grouped into 4 areas: core concepts, architecture and resilience, security and cost and operations and migration. Each one comes with why it is worth asking and what a strong answer contains, so the same question can be scored the same way by different interviewers.

  • Let candidates answer with the provider they know best; test concepts rather than one vendor's product names
  • Use a short design exercise (a web app that must survive a zone failure) and probe trade-offs
  • Include a cost question; candidates who have never looked at a bill are a risk in most cloud roles
For hiring teams

Cohesyve finds, contacts and tests Cloud Computing candidates on the real work, then ranks them for these interviews. 14-day free trial for 10 candidates, not charged until it ends.

Preparing for this interview yourself? Practise it free →

Cloud computing questions come up in interviews for cloud engineers, DevOps engineers, solutions architects and many developer roles. These cloud computing interview questions start with core concepts (service and deployment models, shared responsibility) and move to architecture, security, cost and operations. They are written to be provider-neutral, with AWS, Azure or Google Cloud examples welcome in answers. Each question includes what a strong answer contains, for interviewers scoring and candidates preparing.

Core Concepts

The vocabulary every cloud candidate should be able to explain clearly.

  1. 1

    How would you define cloud computing?

    Why ask it

    Tests whether the candidate can explain the basics precisely.

    What to look for

    On-demand network access to a shared pool of configurable computing resources that can be provisioned and released quickly with minimal management effort, close to NIST's definition. Strong answers name the essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service.
  2. 2

    Explain IaaS, PaaS and SaaS with an example of each.

    Why ask it

    A standard question that reveals clarity of thinking.

    What to look for

    IaaS provides compute, storage and networking where the customer manages operating systems and applications (virtual machines). PaaS lets the customer deploy applications onto a managed platform without managing the servers or OS (managed app platforms). SaaS delivers a finished application (web email, CRM). Good answers explain what the customer controls in each.
  3. 3

    What are the main cloud deployment models?

    Why ask it

    Basic architecture vocabulary.

    What to look for

    Public, private, community and hybrid clouds, as NIST lists them, with an example of when each fits. Many candidates will also discuss multi-cloud; strong ones explain its real costs in skills and complexity.
  4. 4

    What is the shared responsibility model?

    Why ask it

    Misunderstanding it causes security incidents.

    What to look for

    The provider secures the cloud itself (hardware, facilities, the infrastructure that runs services); the customer secures what they put in it (data, identity and access, configuration). The split shifts by service: on virtual machines the customer patches the guest OS; on managed services the provider handles more, but the customer still owns data, encryption settings and permissions.
  5. 5

    What is the difference between scalability and elasticity?

    Why ask it

    Two terms often used interchangeably.

    What to look for

    Scalability is the ability to handle more load by adding resources (vertical or horizontal). Elasticity is adding and removing resources automatically as demand changes, so you pay for what you use. Examples with auto-scaling groups or serverless functions.
For hiring teams

Before the Cloud Computing interviews, see who can do the work

Cohesyve finds and contacts Cloud Computing candidates from your job description and tests each one on an AI job simulation of the role, so these questions go to a ranked shortlist. 14-day free trial for 10 candidates, not charged until it ends.

Architecture and Resilience

How the candidate designs systems that stay up and recover from failure.

  1. 6

    How would you design a web application to survive the failure of a single data centre?

    Why ask it

    High availability is central to cloud architecture.

    What to look for

    Running across multiple availability zones, a load balancer in front of stateless application servers, a managed database with a standby in another zone, health checks and automatic replacement, and no single points of failure. Strong answers mention testing failover.
  2. 7

    What are RTO and RPO, and how do they shape a disaster recovery plan?

    Why ask it

    Tests business-aware resilience thinking.

    What to look for

    RTO is how quickly a service must be restored; RPO is how much data loss is acceptable, measured in time. Tighter targets need more expensive designs, from backup and restore up to active-active across regions. Good candidates start from business requirements.
  3. 8

    When would you choose containers, virtual machines or serverless functions?

    Why ask it

    Shows judgement across compute options.

    What to look for

    VMs for full OS control or legacy software; containers for portable, consistent services and microservices, often with an orchestrator; serverless for event-driven or spiky workloads where you do not want to manage servers. Trade-offs include cold starts, execution limits, cost at steady load and team skills.
  4. 9

    What frameworks or principles do you use to review a cloud architecture?

    Why ask it

    Shows whether the candidate reviews designs systematically.

    What to look for

    A structured review such as AWS's Well-Architected Framework, whose six pillars are operational excellence, security, reliability, performance efficiency, cost optimisation and sustainability, or the equivalent guidance from other providers. The value is in the questions asked, not in naming the framework.

Security and Cost

Two areas where cloud mistakes are expensive.

  1. 10

    How do you apply least privilege in a cloud environment?

    Why ask it

    Identity is the main security boundary in the cloud.

    What to look for

    Role-based access with narrowly scoped permissions, temporary credentials instead of long-lived keys, separate accounts or projects per environment, multi-factor authentication for people, regular access reviews and logging of who did what.
  2. 11

    A storage bucket containing customer data was found publicly accessible. What do you do?

    Why ask it

    A realistic incident scenario.

    What to look for

    Remove public access immediately, check access logs to see what was accessed, follow the incident and data-breach process (including legal and notification duties), find how it happened, and add guardrails such as account-level public access blocks and policy checks in deployment pipelines.
  3. 12

    Cloud costs have risen sharply this quarter. How do you investigate and reduce them?

    Why ask it

    Cost management is part of most cloud roles.

    What to look for

    Breaking down spend by service and tag, finding idle or oversized resources, rightsizing, scheduling non-production environments, committing to discounted capacity for steady workloads, reviewing storage tiers and data transfer, and setting budgets and alerts. Tagging discipline makes this possible.
  4. 13

    How do you protect data at rest and in transit?

    Why ask it

    Basic but essential.

    What to look for

    Encryption at rest using provider-managed or customer-managed keys, TLS for data in transit, key management and rotation, restricting who can use keys, and classifying data so the most sensitive gets the strongest controls.

Operations and Migration

How the candidate runs and moves workloads in the cloud.

  1. 14

    What is infrastructure as code, and why use it?

    Why ask it

    Modern cloud teams manage infrastructure through code.

    What to look for

    Defining infrastructure in version-controlled files (Terraform, CloudFormation, Bicep or similar) so environments are repeatable, reviewed and auditable. Strong answers mention state management, drift detection and running changes through CI/CD.
  2. 15

    How would you approach migrating an on-premises application to the cloud?

    Why ask it

    Migrations are a common project.

    What to look for

    Discovery of dependencies and data, choosing a strategy per application (move as-is, re-platform onto managed services, or re-architect), a pilot, data migration and cutover plans, security and compliance review, and post-migration optimisation. Not every application should move.
  3. 16

    What do you monitor in a cloud application, and how do you alert on it?

    Why ask it

    Tests operational maturity.

    What to look for

    Metrics (latency, error rate, traffic, saturation), logs, traces, cost and security events. Alerts tied to user impact and service-level objectives rather than every metric, with runbooks so on-call engineers know what to do.

Running the interview well

  • 1Let candidates answer with the provider they know best; test concepts rather than one vendor's product names
  • 2Use a short design exercise (a web app that must survive a zone failure) and probe trade-offs
  • 3Include a cost question; candidates who have never looked at a bill are a risk in most cloud roles
  • 4Ask about a real incident they handled and what they changed afterwards
  • 5Check security thinking with a scenario rather than a definition
  • 6Treat cloud certifications as useful evidence of study, and weigh hands-on project experience above them

Common questions

What are the basic cloud computing interview questions?

Expect questions on the definition of cloud computing, IaaS, PaaS and SaaS, deployment models, the shared responsibility model, scalability versus elasticity, availability zones and regions, and basic security such as identity and encryption.

What are the five characteristics of cloud computing?

NIST's definition lists on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service.

Which cloud provider should I learn for interviews?

Learn the one the employer uses if you know it; otherwise pick one major provider and learn it well. The core concepts (compute, storage, networking, identity, scaling, monitoring) carry across providers, and interviewers often care more about the concepts than the product names.

How should I prepare for a cloud computing interview?

Review core concepts, then build something small: a web app behind a load balancer with a managed database, deployed with infrastructure as code and monitored. Be ready to explain the design, its costs, how it fails and how you would secure it.

What is the shared responsibility model in cloud computing?

It divides security duties between the provider and the customer. The provider secures the underlying infrastructure; the customer secures their data, identities, access permissions and configuration. How much the customer manages depends on the service type.

For hiring teams

Save these questions for the Cloud Computing candidates who can do the work

Cohesyve takes your Cloud Computing job description, finds candidates, reaches out to them and runs an AI job simulation of the real work, then returns a ranked shortlist.

1,500+

assessments completed

50%

faster time-to-hire

5 min

from JD to assessment

14-day free trial · 10 candidates · Not charged until the trial ends

For job seekers

Preparing for a Cloud Computing interview yourself? Practise on an AI job simulation of the role. 3-day free trial, cancel anytime.

Practise a Cloud Computing interview free →

See Cohesyve in action

Free 30-min walkthrough

See it on your role