Common Questions & Scenarios
Most "tell me about a time" questions are the same five intents in different costumes. If you prepare the intent, you can reuse a story instead of inventing a new one on the spot.
Goal: show, do not claim. "I am a great collaborator" is empty. A disagreement, what you did, and the outcome is a signal.
[!TIP] ELI5: The exam has a rubric They are not collecting fun anecdotes. Each question is a box on a score sheet: conflict, failure, initiative, influence, motivation. Your job is to drop evidence in the box, then stop.
1. Conflict with a coworker
Intent: Can you disagree without becoming the problem?
Wrong: Never had a conflict. Or: they were an idiot, I won.
Right: A professional disagreement (API shape, launch date, on-call load). You listened, you brought data, you changed your mind or you committed after being overruled.
Traits: empathy, specificity, no character assassination.
Follow-ups they will ask: "What did they think of you after?" "Would you do the same again?"
Sketch: "I wanted a sync REST call; they wanted a queue. I was wrong about peak load. We shipped the queue; I wrote the consumer. p99 stayed under 200ms on Black Friday."
2. Greatest failure
Intent: Do you own mistakes, or do you narrate other people's?
Wrong: "I'm a perfectionist." Or a failure with no change afterward.
Right: You caused it (or a large share). You mitigated. You installed a system so it cannot recur (test, alert, checklist, design review).
Traits: ownership, calibration, no self-flagellation theater.
Never pick a story that makes you look careless with user data unless you can show a serious process change. Interviewers still have to sleep at night.
3. Took initiative / leadership without title
Intent: Do you move the team, or only your ticket?
Wrong: Doing your assigned Jira well.
Right: A problem that was nobody's job: flaky CI, missing runbook, a junior drowning, a cost spike. You aligned people, you shipped the fix, you left it owned.
Traits: scope, influence, finishing.
Initiative that created extra work for everyone with no buy-in scores negative. Leadership includes "I stopped after two people said no, and I was right to."
4. Difficult stakeholder
Intent: Influence without authority. Product, sales, another team, a founder.
Wrong: "PMs don't understand engineering."
Right: Their incentive (launch date, revenue, a customer). You made the trade-off visible (risk, hours, quality). You offered options, not a lecture. You stayed aligned after the decision.
Traits: communication, negotiation, respect for their metric.
5. Why this company?
Intent: Did you do homework, or is this spray-and-pray?
Wrong: Comp, brand, "you're a market leader."
Right: One product decision, one engineering blog, one problem they are clearly solving — tied to what you want to work on next. Two minutes of research beats a paragraph of flattery.
If you cannot name a product surface, you are not ready for that onsite.
6. Ambiguity ("not enough requirements")
Intent: Can you ship without a perfect spec?
Show: you listed unknowns, you picked a default, you time-boxed a spike, you confirmed with the stakeholder, you documented the decision. "I waited for a perfect PRD" is a no-hire for senior roles.
Story bank grid
Reuse. One outage story can be technical depth or failure or leadership depending on which STAR slice you emphasize.
| Story | Conflict | Failure | Leadership | Technical |
|---|---|---|---|---|
| Payment gateway TLS outage | X | X | ||
| Mentoring a junior through a bad review | X | |||
| API design disagreement | X | X | ||
| Migration that locked prod | X | X | ||
| Stakeholder wanted to skip rate limits | X | X |
Write one sentence per filled cell from your own career before the interview week. If a cell is empty, you have a gap; go find a smaller real story, do not invent a war.
[!NOTE] When they ask a question you already used a story for, say so and offer a second: "I used the TLS outage for failure; for conflict I have the API review." Repeating the same hero narrative three times looks like a thin career.
What to remember
- Map questions to intents, then to stories, not to memorized scripts.
- Strong answers include a trade-off, your actions, and a result someone else could verify.
- Empty cells in the grid are practice homework, not a mystery on the day.