Why behavioral questions matter
Technical skills get you in the room. Behavioral answers decide if you get the offer.
Companies ask behavioral questions because past behavior is the best predictor of future behavior. They’re not looking for perfect stories — they’re looking for evidence that you can operate at the level the role requires. A weak technical candidate with sharp behavioral answers often beats a strong technical candidate who can’t explain what they’ve actually done.
The STAR framework
S — Situation: Set the context. Where were you, what was the team, what was the constraint?
T — Task: What specifically was your responsibility? Not the team’s job — yours.
A — Action: What did you do? This is the meat. Be specific. Name the technique, the trade-off, the decision. No passive voice (“we decided to” → “I proposed that we”).
R — Result: What happened? Numbers > adjectives. “Reduced p95 latency from 800ms to 120ms” beats “significantly improved performance”.
The ratio that matters
Most candidates spend 70% on Situation, 20% on Action, 10% on Result.
Flip it: 10% Situation, 60% Action, 30% Result.
Interviewers already know what a startup is. They don’t need three minutes of context. They need to know what you did and what changed because of it.
Building your story bank
Before interviews, prepare 6–8 stories that each cover multiple question types. You want depth, not breadth — one rich story beats five shallow ones.
Suggested stories to prepare:
| Story | Covers |
|---|---|
| A project you led end-to-end | Leadership, ownership, delivery |
| A technical decision you pushed back on | Conflict, conviction, technical judgment |
| A time you failed or made a mistake | Self-awareness, learning, resilience |
| A time you unblocked a team or person | Collaboration, influence, initiative |
| A time you worked under ambiguity or changing requirements | Adaptability, prioritization |
| A time you improved a process or system nobody asked you to | Ownership, proactivity |
| A time you disagreed with your manager | Courage, communication |
| A time you mentored or grew someone | Leadership, communication |
For each, write out the full STAR in writing first. Saying it out loud feels different from having it in your head.
Example question and answer
Question: “Tell me about a time you had to make a technical decision with incomplete information.”
Weak answer: “We had this situation where we needed to choose a database but didn’t have all the requirements. We evaluated a few options and went with PostgreSQL. It worked out well.”
Strong answer (STAR):
Situation (brief): “At my previous company we were migrating a legacy monolith to microservices. One service owned all payment records — about 200M rows — and we had three weeks to decide the new datastore before the team was blocked.”
Task: “I was the only senior engineer on the payment team at the time, so the call was mine to justify.”
Action (detailed): “We had contradictory requirements: the product team wanted flexible querying for a new analytics feature, finance needed ACID guarantees for reconciliation, and infra wanted minimal ops overhead. I mapped out the non-negotiables — ACID was hard, analytics was soft — and ran a load test against PostgreSQL with a read replica versus DynamoDB with DynamoDB Streams. PostgreSQL hit our 99th-percentile latency targets at projected scale; DynamoDB didn’t without significant denormalization that would have broken the analytics queries. I wrote a one-page ADR with the results and got sign-off in two days.”
Result: “We shipped on schedule. Eighteen months later the payment service is handling 3× the original load on the same stack with no incidents. The ADR template I wrote became the company standard for infrastructure decisions.”
Calibration: what level are they assessing?
The same question gets a different expected answer at different levels.
| Level | What they’re listening for |
|---|---|
| L3/SWE I–II | You understood the problem, you executed, you learned |
| L4/SWE III | You drove decisions, you handled ambiguity, you influenced peers |
| L5/Senior | You owned outcomes, you shaped the problem, you multiplied the team |
| L6/Staff+ | You changed direction of a product or team, you operated across orgs |
Before answering, know what level you’re interviewing for and calibrate your action and result accordingly.
The tell for a weak story
You’re in trouble if you catch yourself saying:
- “We” without ever specifying what you did
- “I think it went well” (no data)
- “It was a team effort” (deflecting ownership)
- A story where nothing actually went wrong (no adversity = no interesting signal)
Interviewers probe weak stories with follow-ups. Strong stories survive follow-ups because the detail is real.