Site Reliability Engineer
Interview Questions.
15 practice questions across 13 skills — each with what the interviewer is really listening for. Prepare with intent, not guesswork.
Free · No signup · Grouped by skill
How to use this page
Don't memorise answers. For each question, read what the interviewer is looking for, then practise answering out loud in your own words with a real example. The notes describe strong answers and common red flags — they are for your prep, not a script.
SLOs
1 questionWhat are SLIs, SLOs, and error budgets, and how do they change how you work?
AdvancedWhat the interviewer is looking for: Strong answers define indicators, targets, and the budget for unreliability, and use it to balance reliability against feature velocity. Chasing 100% uptime with no error budget is a red flag.
Incident Management
1 questionA service is down at 3am and you are on call. Walk me through your first 15 minutes.
AdvancedWhat the interviewer is looking for: Look for assessing impact, mitigating first (rollback, failover), communicating status, and then finding root cause. Debugging deeply while users suffer instead of mitigating is a red flag.
Monitoring
1 questionHow do you decide what to monitor and alert on so on-call is sustainable?
AdvancedWhat the interviewer is looking for: Strong candidates alert on user-facing symptoms and the golden signals, make alerts actionable, and cut noise. Alerting on everything until people ignore pages is a red flag.
Kubernetes
1 questionHow do you debug high latency on a service running in Kubernetes?
AdvancedWhat the interviewer is looking for: Look for checking resource limits/throttling, pod health, network, dependencies, and tracing, systematically. Restarting pods blindly is a red flag.
Linux
1 questionHow do you diagnose a Linux server that is slow but not obviously broken?
IntermediateWhat the interviewer is looking for: Strong answers use tools for CPU, memory, disk I/O, and network to isolate the resource, forming hypotheses. Random restarts with no diagnosis are a red flag.
Python
1 questionHow do you use Python or scripting to reduce toil?
IntermediateWhat the interviewer is looking for: Look for automating repetitive operational work and building tooling, treating toil as something to engineer away. Accepting endless manual work is a red flag for SRE.
CI/CD
1 questionHow do you design a CI/CD pipeline so a bad deploy does not cause an outage?
AdvancedWhat the interviewer is looking for: Strong candidates use automated tests, canary or progressive rollouts, health checks, and fast rollback. Deploying 100% at once with no rollback is a red flag.
Blameless Postmortems
1 questionWhat makes a good blameless post-mortem, and why blameless?
IntermediateWhat the interviewer is looking for: Look for focusing on systemic causes and action items, encouraging honesty by not blaming individuals. Pointing fingers at a person is a red flag that kills learning.
On-Call Discipline
1 questionHow do you keep on-call healthy and prevent burnout on the team?
IntermediateWhat the interviewer is looking for: Strong answers reduce alert noise, fix recurring issues, share load fairly, and follow up on toil. Accepting a noisy, burnout-inducing rotation is a red flag.
Terraform
1 questionHow do you use Infrastructure as Code with Terraform to make systems reproducible?
IntermediateWhat the interviewer is looking for: Look for declarative, version-controlled, reviewed infra with managed state, and no manual drift. Hand-editing infra alongside Terraform is a red flag.
Problem Solving
1 questionHow do you approach capacity planning so you neither waste money nor fall over under load?
AdvancedWhat the interviewer is looking for: Strong candidates use metrics, load testing, and autoscaling with headroom for spikes. No capacity planning until something breaks is a red flag.
Working Under Pressure
1 questionHow do you stay effective and clear-headed during a high-pressure outage?
IntermediateWhat the interviewer is looking for: Look for following the incident process, clear roles and communication, and methodical action. Panic or heroic solo cowboy fixes are red flags.
Prometheus
1 questionHow do you use Prometheus and Grafana to make a service observable?
IntermediateWhat the interviewer is looking for: Look for meaningful metrics, dashboards around the golden signals, and alerting tied to SLOs. Dashboards nobody uses or vanity metrics are a red flag.
Behavioural & role-general
2 questionsTell me about a major incident you handled. What systemic change came out of it?
BasicWhat the interviewer is looking for: Strong answers show calm response and a real prevention improvement, not just a quick patch. Blaming a person or claiming perfect uptime are red flags.
How do you balance reliability work against pressure to ship features?
IntermediateWhat the interviewer is looking for: Look for using error budgets and data to make the trade-off objective rather than emotional. Always deferring reliability for features is a red flag.