Separate the responsibilities in the role
Describe the cloud platforms, major workloads and current operating model. Explain whether the priority is migration, platform improvement, day-to-day support or solution design. Identify the teams responsible for applications, security and infrastructure, including external providers.
State what the candidate will own and which decisions need approval. An architect, cloud engineer and administrator may contribute differently even when they use similar platform names. If you expect one person to cover several functions, be explicit about the priorities and the support available.
Use scenarios that reveal trade-offs
Ask the candidate to walk through a relevant incident or change. What did they check first, how did they assess risk and how did they communicate with the business? Explore rollback, monitoring and how they knew the change had worked.
Microsoft’s Azure Well-Architected Framework provides useful discussion areas including reliability, security, cost optimisation, operational excellence and performance efficiency. Use those themes to frame questions where relevant, not as a claim that one checklist certifies a candidate’s competence. The scenario should reflect your workloads and be reviewed by someone qualified to assess the answers.
Check the operating habits
Discuss access control, documentation, automation and handover at a level appropriate to the job. Ask how the candidate avoids undocumented manual changes and how they keep colleagues informed. For senior roles, explore how they decide what not to build and when additional complexity is justified.
Be clear about on-call arrangements, support coverage and authority during an incident. A role that includes out-of-hours responsibility needs a practical conversation about the arrangement. Verify relevant credentials where required, while keeping delivery examples and judgement central to the decision.
PUT IT INTO PRACTICE
A cloud interview prompt
“A business application is experiencing intermittent failures after a change. Users need an update, and the cause is not yet known. Explain your first steps, who you would involve and how you would decide between further investigation and rollback.” Assess the reasoning and assumptions, rather than expecting a universal technical answer.
Your practical checklist
- Define architecture, delivery and operational ownership.
- Describe the workloads and supporting teams.
- Assess change, incident and recovery reasoning.
- Clarify support coverage and escalation authority.
Take it forward
The strongest cloud shortlist connects platform experience to the responsibilities of your environment. Look for clear reasoning about safe operation as well as familiarity with tools.
