Behavioral interview questions
Behavioral interview questions ask candidates to describe something they actually did — "tell me about a time…" — because past behavior predicts future behavior better than hypotheticals. Pick five or six of the 13 below and push each answer to a real situation, the candidate's own actions, and a verifiable result.
All 13 questions
Common behavioral interview questions
- 01Describe a project you're proud of. What was your specific contribution?
- 02Tell me about a time you disagreed with your manager and said so.
- 03Describe a deadline you missed. What did you do when you saw it slipping?
- 04Describe a time you had to work closely with someone whose style clashed with yours.
- 05Tell me about the toughest piece of feedback you've received. What did you do with it?
Behavioral questions about working with AI
- 06Tell me about a time you used an AI tool to get a piece of work done faster. How did you check the output?
- 07Describe a time an AI tool got something wrong in your work. How did you catch it, and what changed afterwards?
- 08What in your work have you deliberately kept doing yourself, even though an AI tool could do it faster?
- 09How has the way you work changed in the last two years as AI tools improved? Walk me through a before-and-after.
Behavioral questions for pressure and ambiguity
- 10Tell me about your biggest professional failure.
- 11Tell me about a time you had too many priorities and had to drop something. How did you choose?
- 12Describe a decision you made with incomplete information. How did you decide, and how did it turn out?
- 13Tell me about a time you noticed a problem nobody owned and dealt with it.
Questions 1–5
Common behavioral interview questions
Describe a project you're proud of. What was your specific contribution?
A strong answer isolates the candidate's own decisions inside the team result, gives accurate credit to others, and explains why this project mattered to them — honest 'I' and honest 'we' in the same story.
What to look for
- Can isolate their own decisions and actions inside the team result.
- Gives credit accurately without diminishing their own part.
- Explains why this project, which reveals what they value.
Example answerRed flags
Example answer
We rebuilt onboarding and doubled activation. My part was the diagnosis: I ran twenty user session recordings, found that people stalled on the workspace-setup step, and made the case to cut it entirely. Design and engineering built the new flow — that wasn't me — but the decision about what to remove was mine, and it's what moved the number.
Red flags
- Every sentence is 'we' and no follow-up can extract what they personally did.
- Takes credit for outcomes their role couldn't plausibly have driven.
Tell me about a time you disagreed with your manager and said so.
Good answers raise the disagreement to the manager's face, argue from evidence, and commit genuinely to the final call — with an honest account of how it turned out, whichever way it went.
What to look for
- Disagreed to the manager's face, not around them.
- Argued from evidence or user impact, not preference.
- Committed to the decision afterwards, even when overruled.
Example answerRed flags
Example answer
My manager wanted to ship a feature behind a waitlist to build hype. I thought it would kill our activation experiment and said so in our one-on-one, with numbers from the last launch. She shortened the waitlist to a week. She was right about the hype, I was right about the dip — we got both.
Red flags
- Has never disagreed with a manager, or frames every past manager as an obstacle.
- Was overruled and quietly worked against the decision.
Describe a deadline you missed. What did you do when you saw it slipping?
Strong candidates flag the slip early, arrive with a revised plan — cut scope, new date, added help — and change how they estimate afterwards. Weak ones discover the miss at the deadline, with everyone else.
What to look for
- Raised the flag when the slip became likely, not certain.
- Brought options, not just apologies: cut scope, moved the date, added help.
- Changed something in how they estimate or commit afterwards.
Example answerRed flags
Example answer
A client report was due Friday; on Tuesday I realized data cleanup alone would eat two more days. I called that afternoon and offered topline numbers Friday, full report Wednesday — what they actually needed Friday was one slide for their board. Now I split every deliverable into the part needed on the date and the part that can follow.
Red flags
- The stakeholder learned about the miss at the deadline.
- The story blames scope, tools, or teammates with no part owned.
Describe a time you had to work closely with someone whose style clashed with yours.
Strong answers describe the other person's style as a fact to work with, name what the candidate changed in their own behavior, and say honestly whether the relationship improved.
What to look for
- Describes the other person's style neutrally, without a villain edit.
- Adapted their own behavior: channel, cadence, or level of detail.
- The relationship ended better than it started, or says honestly why not.
Example answerRed flags
Example answer
I think out loud in meetings; a designer I worked with needed everything in writing a day ahead. For a month we frustrated each other. The fix was embarrassingly small: I sent a five-line agenda each evening, she flagged what she wanted to talk through live. Her pushback got sharper and our shipped work got better.
Red flags
- The entire story is about what was wrong with the other person.
- The resolution was avoiding the person or escalating to remove them.
Tell me about the toughest piece of feedback you've received. What did you do with it?
A strong answer quotes the feedback nearly verbatim, admits it stung, and points to a durable behavior change you could verify — the habit that exists because of it.
What to look for
- Quotes the feedback close to verbatim — vagueness means it bounced off.
- An emotional beat that rings true: good candidates admit it stung.
- A durable, checkable behavior change, not a promise to 'work on it'.
Example answerRed flags
Example answer
A peer told me in a retro that when I'm ahead of the room I get impatient, so people had stopped raising half-formed ideas around me. That one hurt because I saw it immediately. I started asking one question before offering any opinion, and eight months later the same peer said the room felt different.
Red flags
- The toughest feedback they can produce is a disguised strength ('too much of a perfectionist').
- Dismisses the feedback as the giver's problem.
Questions 6–9
Behavioral questions about working with AI
2026 · AITell me about a time you used an AI tool to get a piece of work done faster. How did you check the output?
A strong answer names a specific task, the time actually saved, and a verification step proportionate to the risk. Unchecked AI output is a quality risk; refusing AI outright is a speed risk.
What to look for
- A specific task with real stakes, honest accounting of time saved.
- Verification proportionate to risk — spot-checks low stakes, full review customer-facing.
- Knows what the tool is bad at, from experience.
Example answerRed flags
Example answer
I use an LLM to draft our monthly changelog from release notes — two hours down to twenty minutes. But the twenty minutes are the job: I check every claim against what actually shipped, because it once called a flagged feature 'available to all users' and a support ticket taught me that. Draft with the tool, own every word.
Red flags
- Can't name a verification step for anything they've shipped with AI help.
- Refuses AI tools on principle without having tried them for anything.
Describe a time an AI tool got something wrong in your work. How did you catch it, and what changed afterwards?
Strong answers name a specific error, show it was caught by a process — review habit, test, second source — and end with a workflow change matched to that failure mode. 'It's never been wrong' means they weren't checking.
What to look for
- A specific, plausible error: invented figure, wrong API call, confident mis-summary.
- Caught by their process, not by luck or a customer.
- A concrete workflow change followed, matched to the failure mode.
Example answerRed flags
Example answer
I asked an assistant to summarize five customer interviews and it attributed a churn reason to the wrong segment — plausible-sounding, completely wrong. I caught it because I'd been in three of the calls. Now every claim in a summary links back to its source quote, and anything I wasn't present for gets checked against the transcript.
Red flags
- Claims AI tools have never produced an error in their work.
- Found the error and kept the same unchecked workflow anyway.
What in your work have you deliberately kept doing yourself, even though an AI tool could do it faster?
The strong answer names a deliberate choice with a defensible reason — skill maintenance, trust, quality of thought — and reveals what the candidate believes their own judgment is for.
What to look for
- A deliberate choice with a defensible reason, not vague preference.
- Distinguishes 'AI can't do this well' from 'AI shouldn't do this'.
- Consistent with their other answers about where they do use AI.
Example answerRed flags
Example answer
Two things. I write performance feedback myself — the person can tell, and the thinking I do while writing is most of the value. And I take my first pass of user-interview notes by hand, because choosing what to write down is where I notice patterns. I'll let a model transcribe and challenge my summary. My read first, then the tool's.
Red flags
- Nothing is reserved — they'd delegate any task to AI if it were faster.
- The reserved list is everything, which usually signals discomfort rather than judgment.
How has the way you work changed in the last two years as AI tools improved? Walk me through a before-and-after.
A strong answer is a concrete before-and-after on a real recurring task: the old workflow, the new one, what got better, and the new failure mode they now guard against.
What to look for
- A specific recurring task, with old and new workflow both described.
- Deliberate adoption: evaluated, kept what worked, dropped what didn't.
- Names a new failure mode and the guard against it.
Example answerRed flags
Example answer
Two years ago research meant eight browser tabs and a full day. Now a deep-research run maps the terrain in an hour and I spend the afternoon verifying the three claims that actually matter. Half the time, more ground. The trap is fluency — a polished summary feels finished, so I check the load-bearing facts before anything ships.
Red flags
- Nothing about how they work has changed since 2023.
- Adopted tools everywhere with no story about what got worse or needed guarding.
Questions 10–13
Behavioral questions for pressure and ambiguity
Tell me about your biggest professional failure.
The best answers pick a failure with real stakes — money, a launch, trust — own the candidate's causal part precisely, and name the specific practice that exists because of it.
What to look for
- The failure is real — money lost, launch missed, trust damaged.
- Causal honesty: what they did or didn't do, separate from luck.
- A changed practice you could verify with their next team.
Example answerRed flags
Example answer
I hired the wrong person and kept them nine months past the evidence. I'd skipped reference checks because a friend recommended them, then kept explaining away misses because firing my friend's referral felt awful. Now I do references on every hire, and I write down in advance what 'working out' looks like at ninety days.
Red flags
- Offers a failure with no stakes, or one engineered to showcase a strength.
- The autopsy assigns every cause to circumstances.
Tell me about a time you had too many priorities and had to drop something. How did you choose?
A credible answer names what was actually dropped, ties the choice to impact rather than the loudest stakeholder, and shows the affected people heard about it before they found out on their own.
What to look for
- Something real got dropped or shrunk, with a named cost.
- Reasoning tied to impact, not whichever stakeholder shouted loudest.
- Told whoever was counting on the dropped work, proactively.
Example answerRed flags
Example answer
During a product launch I was also running booth prep and a quarterly report. I cut the report to a two-page summary — its audience needed a decision, not a document — and told the exec two weeks ahead. She took the summary and never asked for the long version again, which taught me something about that report.
Red flags
- Claims they got everything done by working harder — nothing was ever dropped.
- Dropped something silently and let the owner discover it.
Describe a decision you made with incomplete information. How did you decide, and how did it turn out?
Good answers separate what could be learned cheaply from what couldn't, size the call to its reversibility, and report the actual outcome — including when the bet was wrong.
What to look for
- Distinguished what could be learned cheaply from what couldn't in time.
- Weighed reversibility: fast on two-way doors, slower on one-way doors.
- Reports the actual outcome, including if the call was wrong.
Example answerRed flags
Example answer
We had a week to accept or decline a partnership slot — no time for real due diligence. I spent two days checking references and their churn story, and took the strategic fit as a bet because the exit clause made it reversible in a quarter. It turned out mediocre; we exited at the clause. Same information, same bet again.
Red flags
- Postponed the decision until it made itself.
- Retrofits certainty — claims they knew it would work all along.
Tell me about a time you noticed a problem nobody owned and dealt with it.
Strong answers show the candidate spotted the gap themselves, fixed or escalated it properly, and gave it a durable owner afterwards instead of quietly becoming that owner forever.
What to look for
- Noticed the problem themselves rather than being assigned it.
- Fixed it or routed it to the right owner — either works.
- Closed the loop: the gap has a durable owner or process now.
Example answerRed flags
Example answer
Customer emails about invoices bounced between support and finance for a week each; nobody owned billing questions. I wrote a one-page triage guide with finance, got both teams to agree, and set up a shared inbox rule. Response time dropped to a day. Then I made it finance's process, not my hobby — unofficial ownership is how people burn out.
Red flags
- Only notices problems; every story ends with 'I raised it' and nothing after.
- Hoards ownership — became the bottleneck for everything they touched.
Scoring rubric
| Score | Evidence anchor |
|---|---|
| 1 | Speaks in hypotheticals and generalities — 'I would…', 'I always…' — and no follow-up produces a real situation with their own actions in it. |
| 2 | Produces real examples but stays on the surface: team outcomes claimed as personal ones, no numbers, failures framed as circumstances. |
| 3 | Concrete situations with clear personal actions and honest results, including at least one owned failure. Some answers lack reflection on what changed afterwards. |
| 4 | Specific, verifiable stories across the set: owns causes precisely, names what changed in their practice, and shows working judgment about AI in their own workflow. |
| 5 | Stories carry dates, numbers, and named trade-offs; the candidate volunteers the unflattering detail before you dig for it, and every scar comes with the habit it produced. You end the hour knowing exactly how they operate. |
Frequently asked questions
What are behavioral interview questions?
Behavioral interview questions — also called behavioral-based interview questions — ask a candidate to describe a specific past situation and what they personally did in it. Past behavior under real constraints predicts future behavior better than rehearsed intentions.
What's the difference between behavioral and situational interview questions?
Behavioral questions ask about something that actually happened; situational questions pose a hypothetical and ask what the candidate would do. Behavioral answers are harder to fake and easier to probe.
What are the most common behavioral interview questions?
Lists of top behavioral interview questions converge on the same five: failure, conflict with a colleague, missed deadline, disagreement with a manager, proudest achievement. All five are in the set above; the AI section covers the ones that aren't common yet but should be.
How should candidates answer behavioral interview questions?
The standard advice is the STAR method: situation, task, action, result. As the interviewer, use it as your probe — push on the result and what changed afterwards. That's where prepared stories run out.
How many behavioral questions should one interview include?
Five or six, chosen for the competencies the role needs, beat all 13 rushed — a good behavioral answer takes three to five minutes with follow-ups. Kira can cover the wider set by voice before the human round.