DevOps / SRE interview prep, head-to-head
Generic copilots have not done a 3-AM rollback. The DevOps copilot has indexed the patterns the senior SRE interviewer expects you to know cold.
Built for the four DevOps / SRE rounds
Kubernetes design rounds
Multi-cluster topology, namespace strategy, network policy, ingress design, GitOps with ArgoCD or Flux, service mesh trade-offs — the copilot anchors the cluster size, then walks the canonical decisions a senior SRE is expected to make.
IaC + drift rounds
Terraform vs Pulumi vs CloudFormation, monolithic state vs workspace split, drift detection, modules-vs-stacks, secret management, plan-review automation. The copilot covers IaC at the trade-off layer, not at the syntax layer.
Incident-response walkthroughs
Scenario: "API latency spiked at 14:00 UTC, oncall paged, what do you do?" The copilot scaffolds the IR pattern — confirm symptom, freeze deploy, triage by recent change, capture diagnostic data, communicate, mitigate, post-mortem — with the timing a real on-call follows.
SLO + error-budget rounds
When the interviewer says "propose an SLO for this service", the copilot walks the canonical method: define SLI from the customer's perspective, set the SLO target via historical p99 + business tolerance, define the error budget, define the burn-rate alerts. Acronym fluency loses; this gets you scenario fluency.
CI/CD design rounds
Trunk-based vs gitflow, blue-green vs canary vs progressive delivery, automated rollback strategy, feature-flag-gated rollouts, deploy frequency vs MTTR trade-offs. The copilot has the trade-offs that scale from a 30-person startup to a 5,000-engineer org.
Cloud cost rounds
Right-sizing, reserved capacity vs savings plans vs spot, idle-resource detection, network-egress optimisation, FinOps governance. Senior SRE / platform-engineer loops increasingly include a cost round — the copilot has scaffolds for the five most-asked variants.
