You join a 200-engineer company where every team built its own deployment tooling. What is your strategy?
What they are really testing: Whether you can drive change at scale without mandate-driven failure. They want sequencing, evidence and adoption thinking, not a target architecture slide.
A real interview question
You join a 200-engineer company where every team built its own deployment tooling. What is your strategy?
What most people say
drag me
“Pick the best tooling, standardise on it, and require all teams to migrate by a set date.”
Mandate without adoption is how platform teams fail. Teams that were never asked will comply minimally, keep shadow tooling for what the standard does not cover, and the platform becomes a tax rather than a service.
The follow-ups they ask next
A senior team refuses and has good reasons. What now?
Take the reasons seriously, they are usually a real gap. Either extend the platform to cover it, or agree an explicit exception with a support boundary. A forced migration that degrades a working team costs more than the exception.
How do you know the platform is succeeding?
Voluntary adoption rate, DORA metrics for teams on it versus off it, time from a new service being created to its first production deploy, and whether teams stop building shadow tooling.
What the interviewer is listening for
- Diagnoses why divergence happened first
- Adoption through pull, proven with volunteers
- Realistic multi-quarter timeline
What sinks the answer
- Opens with a mandate
- No migration support
- Promises a fast timeline
If you genuinely do not know
Say this instead of freezing. Reasoning out loud from what you do know beats silence every single time, and a good interviewer is listening for exactly that.
“First [understand why teams diverged, usually the central path was slow or absent]. Then [standardise the common 80% and leave real differences alone], [build a paved road faster than rolling your own], [prove it with 2 or 3 volunteer teams and measure], then [set a decommission date with resourced migration]. Expect [4 to 6 quarters], and say so.”
Keep going with strategy
Principal
When would you tell a company not to use Kubernetes?
Principal
The business wants faster shipping, your on-call team is burning out from incidents. How do you resolve that?
Principal
Your team wants to build an internal developer platform. A vendor sells one. How do you decide?
Principal
Tell me about a time you argued against your own proposal or changed your mind on a significant technical decision.
Principal
One slow dependency takes down your entire system. Why does that happen and how do you prevent it?
Principal
Design multi-region failover for a system that currently runs in one region, targeting 15 minutes recovery.
Knowing the answer is not the same as recalling it under pressure
Sign in to send the questions you fumble to spaced recall, so they come back right before you would forget them, and learn the concepts behind them with hands-on labs.
Start free