Opscraft
Senior DevOps interviews · Lesson 2 of 10 · 5 min read

Your story: introduction and project narratives

The self-introduction (60–90 seconds)

Structure: current role → scope → two or three highlights with impact → strengths → what you want next. A template — replace every [bracket] with your own details:

Enlarge diagramCurrent rolecompany, areaScopestack, scale2–3 highlightswith numbersStrengthswhat you'reknown forNextwhat you want
Figure 1. A 60–90 second introduction in five parts.
Swipe sideways to see all of it, or tap to open it full screen.

"I'm [your name], a DevOps/Platform engineer with about [N] years of experience, currently at [company] working on [area, e.g. platform automation]. My work centres on AWS and Kubernetes — I run workloads on EKS with GitOps using Flux and Kustomize, build CI/CD pipelines in Jenkins and GitHub Actions, and provision infrastructure with CloudFormation, with Maven and Gradle builds on the application side. A couple of things I'm proud of: [e.g. moving N services to a GitOps flow that cut deployment time from X to Y minutes], and [e.g. a cost or reliability improvement with a number]. I'm a CKA and AWS Certified Solutions Architect. Recently I've been strengthening my Terraform skills with an EKS platform project on GitHub. I'm now looking for a senior platform role in Bangalore where I can own larger-scale Kubernetes and cloud infrastructure."

Practise it out loud until it's natural, not memorised word for word. Adjust the emphasis to each JD.

Build an impact inventory before interviewing

For each significant project, write down:

Field Example prompts
Problem What was slow, risky, expensive or manual?
Scale Number of services, clusters, environments, deploys/day, users, ₹/$
Your role What you personally designed, built, decided
Architecture One sentence + a sketch you can draw
Decisions & alternatives Why this approach; what else you considered
Result Numbers: time saved, incidents reduced, cost cut, lead time, MTTR
Lessons What you'd do differently

Numbers matter because they make impact concrete: "cut deploy time from 40 to 12 minutes", "reduced monthly EKS spend by ~25% with Spot and right-sizing", "reduced P1 incidents from deploys from N per quarter to zero with readiness probes and canaries". If you don't have exact numbers, use honest estimates and say so ("roughly", "about").

Explaining a complex project

Use problem → architecture in one sentence → your decisions and trade-offs → result → what's next:

Enlarge diagramProblemslow, risky,manual?Architectureone sentence +sketchYour decisionsand thealternativesResultnumbersNextwhat you'dimprove
Figure 2. Explaining a project: problem first, then a one-sentence architecture, your decisions, and the measurable result.
Swipe sideways to see all of it, or tap to open it full screen.

"Problem: every team deployed to EKS with hand-run kubectl and Jenkins scripts, so environments drifted and rollbacks were manual. Architecture: Git as the source of truth — Kustomize bases and per-environment overlays, Flux reconciling each cluster. My decisions: I chose Flux over Argo CD because [reason], structured overlays per environment, and moved image promotion to PRs so production changes were reviewed. Result: [deploy frequency / rollback time / drift incidents]. Next I'd add progressive delivery with Flagger."

When asked "walk me through your current architecture": one-line summary, then the request path (DNS → ALB → ingress → services → data stores), then the deployment flow (commit → CI → registry → GitOps → cluster), then your specific contributions. Offer to draw it.

"I" and "we"

Use "we" for team outcomes and "I" for your part — clearly separated. "We migrated 40 services to EKS; I designed the ingress and DNS setup and wrote the shared pipeline library." Interviewers are hiring you, not your team.