winkle

Interview engineers without booking a call.

79 ready-made AI interviews that ask, follow up and score. Voice or typed, in 11 languages, and yours to send today.

AI interviewerScored
Example
  1. Interviewer: Tell me about a time you changed someone's mind at work. How did you go about it?

  2. Candidate: Our team wanted to rebuild the billing page from scratch, and I thought it was the wrong call, so we put together a plan to…

  3. Follow-up

    Interviewer: You said "we". What did you say yourself, and what finally changed their mind?

  4. Scored on

    • Ownership30%
    • Structure25%
    • Impact25%
    • Self-awareness20%

    Score, reasons and recording. Your team decides.

An example of how Behavioural interview (STAR) sounds. The real questions open when you use it.

30 interviews · page 2 of 3

  • Security engineer

    Engineering · Technical

    Finds out how a candidate models threats, fixes real weaknesses and handles incidents under pressure.

    Listens for

    • Threat modelling
    • Secure engineering
    • Detection and response
    • Risk judgement and communication
    Voice or typed

    7 of 18 questions · 35 min · Mid to Staff

  • Cloud and platform engineer

    Engineering · Technical

    Finds out how a candidate designs, runs and automates cloud infrastructure that teams can rely on.

    Listens for

    • Infrastructure design
    • Reliability and operations
    • Automation and developer experience
    • Cost and security judgement
    Voice or typed

    7 of 17 questions · 35 min · Mid to Staff

  • Data engineer

    Data · Technical

    Finds out how a candidate builds reliable data pipelines and models that others can trust.

    Listens for

    • Pipeline design
    • Data modelling
    • Reliability and data quality
    • Trade-offs and collaboration
    Voice or typed

    7 of 17 questions · 35 min · Mid to Staff

  • Data scientist

    Data · Technical

    Finds out how a candidate frames problems, models with rigour and proves real-world impact.

    Listens for

    • Problem framing
    • Statistical and modelling rigour
    • Experimentation and causal reasoning
    • Communication and impact
    Voice or typed

    7 of 17 questions · 35 min · Mid to Staff

  • Machine learning engineer

    Data · Technical

    Finds out how a candidate takes models into production and keeps them fast, reliable and accurate.

    Listens for

    • ML system design
    • Production engineering
    • Model quality and monitoring
    • Judgement and trade-offs
    Voice or typed

    7 of 17 questions · 35 min · Mid to Staff

  • Senior product manager: strategy

    Product and design · Product

    Tests whether a senior product leader can set direction, place bets and bring other teams with them.

    Listens for

    • Strategic thinking
    • Market and customer insight
    • Bets and trade-offs
    • Influence and alignment
    • Outcomes and measurement
    Voice or typed

    7 of 18 questions · 35 min · Senior to Staff

  • Sales manager

    Sales · Behavioural

    Finds out whether a sales leader coaches, hires, forecasts and leads a team to its number.

    Listens for

    • Coaching and developing people
    • Forecast and pipeline leadership
    • Hiring and performance management
    • Strategy and influence
    Voice or typed

    7 of 18 questions · 35 min · Senior to Staff

  • Growth and performance marketer

    Marketing · Role-specific

    Tests how a candidate finds, tests and scales channels that bring customers at a sensible cost.

    Listens for

    • Unit economics
    • Experimentation
    • Analysis and attribution
    • Prioritisation and collaboration
    Voice or typed

    7 of 17 questions · 35 min · Mid to Staff

  • Operations manager

    Operations · Role-specific

    Finds out whether a candidate can run reliable processes, lead a team and improve results with data.

    Listens for

    • Process improvement
    • Planning and prioritisation
    • Data and metrics
    • People leadership
    Voice or typed

    7 of 18 questions · 35 min · Mid to Staff

  • Teamwork and collaboration

    Any role · Behavioural

    Finds out how someone contributes to shared work, supports teammates and keeps a group moving.

    Listens for

    • Contribution to shared goals
    • Working across differences
    • Reliability to teammates
    • Generosity and support
    Voice or typed

    7 of 17 questions · 32 min · All levels

  • Finds out whether someone can raise hard things directly, listen under strain and reach a resolution.

    Listens for

    • Directness with respect
    • Listening and understanding
    • Resolution and outcome
    • Composure and self-awareness
    Voice or typed

    7 of 17 questions · 32 min · All levels

  • Resilience under pressure

    Any role · Behavioural

    Finds out how someone prioritises, keeps quality up and recovers when work gets intense or goes wrong.

    Listens for

    • Prioritising under pressure
    • Recovery from setbacks
    • Quality and honesty under strain
    • Learning and prevention
    Voice or typed

    7 of 16 questions · 32 min · All levels

From link to shortlist, while you do something else.

  1. Step 1

    Use an interview

    One click copies it to your account. Swap a question, change the language or the format, or send it as it is.

  2. Step 2

    Send one link

    Candidates take it when it suits them, on any device. The interviewer asks, listens and follows up.

  3. Step 3

    Read the shortlist

    Every answer scored with the reasoning beside it, plus the recording. You decide who goes forward.

Questions, answered.

What employers ask before sending their first AI interview.

  • What happens when I use an interview?

    Winkle copies it into your account as your own interview, with every question, follow-up rule and scoring guide. Change anything you like, then send one link to as many candidates as you want.

  • Who asks the questions?

    An AI interviewer, out loud or in writing. It asks each question in order and, when an answer is vague, asks one follow-up about what the candidate actually said.

  • How are answers scored?

    Every answer is scored against the same guide, with the reasoning written beside the score. You get the recording or the written answers too, so any score can be checked.

  • Does the AI decide who gets hired?

    No. Scores are a recommendation. A person on your team reviews every candidate and makes every decision.

  • Can candidates answer in their own language?

    The interviewer can run the interview in any of 11 languages. Pick the language when you set it up.

  • Why can't I see the questions before I sign up?

    Because candidates could too. Every question opens the moment you use an interview, and you can read, change or replace any of them before it is sent. A candidate who has read the questions in advance is reciting, not answering.

No interview for your role? Winkle writes one.

Name the role or paste the job description. Winkle drafts the questions, the follow-ups and the scoring guide, and you approve every one before it goes out.