How Senior DevOps interviews are structured
Technical knowledge gets you shortlisted; how you explain it gets you the offer. This topic is about the interview itself for Senior DevOps / Platform Engineer roles in Bangalore: what each round evaluates, how to present your five years of AWS/EKS/GitOps work with impact, answer frameworks for technical, design, troubleshooting and behavioural questions, how to handle your known gaps (Terraform, Ansible, Azure) honestly, and the logistics — notice period, CTC conversations and closing. Where you see [brackets], fill in your own real numbers and examples; interviewers can tell rehearsed generic stories from real ones.
Typical rounds at Bangalore product companies and GCCs
| Round | Format | What they're evaluating |
|---|---|---|
| Recruiter / HR screen | 15–30 min call | Fit with the JD, notice period, current and expected CTC, location (and relocation, if it applies), communication |
| Technical 1 — fundamentals | 45–60 min | Linux, networking, Docker, Kubernetes, AWS, CI/CD basics — breadth and accuracy |
| Technical 2 — depth & scenarios | 60 min | Deep dives into your projects, troubleshooting scenarios, "what would you do if..." |
| Hands-on / live (common) | Live coding or take-home | Write a Dockerfile, Kubernetes manifests, a pipeline, a bash/Python script, debug a broken setup, or a Terraform module |
| System design (senior) | 60 min whiteboard | Designing CI/CD, platforms, HA/DR, observability — trade-offs and reasoning |
| Hiring manager / EM | 45–60 min | Ownership, collaboration, handling incidents and ambiguity, growth, why this role |
| HR / bar raiser | 30 min | Culture, motivation, offer discussion |
Not every company has every round; GCCs (global capability centres) of US/European companies often add a values/behavioural round.
What "Senior" changes
Interviewers expect you to:
- Own systems end to end — not just "I wrote the pipeline", but "I designed how we deploy, measured it, and improved it".
- Explain trade-offs — why canary over blue-green, why Karpenter, why Flux, what you'd do differently.
- Lead incidents — calm diagnosis, mitigation first, blameless postmortems, prevention.
- Influence and mentor — reviews, standards, golden paths, helping developers self-serve.
- Think about cost, security and reliability together, not as afterthoughts.
A useful self-check: for every project on your résumé, can you explain the problem, the architecture, your decisions and their alternatives, the measurable outcome, and what you'd improve?