Why do you recommend your AI interview trainer?

The conflict of interest, first

This page answers a question people put to us, and we are not a neutral party. SwissJobs.app earns money from its subscription and the practice interview is part of it. An honest answer has to say that up front and then offer more than a sales pitch: reasons you can test, and limits we name ourselves.

Testing takes a few minutes. Without an account you set up a round, the tool writes two questions about your advert, and you answer the first out loud and get it scored. That is enough to check the three main claims on this page: whether the question really fits your application, what speaking against the clock feels like, and whether the reasoning behind the score is any use. If one of those falls flat, trust the tool over this text.

Reason 1: the questions know your CV and the advert

Lists of common interview questions run into the thousands and share one flaw: they know nothing about you. A prepared hiring manager in Switzerland has your dossier open and asks about the gap in your CV, the move from specialist to team lead, the requirement in the advert you only half meet. Those are the uncomfortable questions, and they are exactly the ones a generic list never makes you rehearse.

So the tool takes two inputs: your CV, uploaded, picked from the CVs you have saved, or pasted, and the job advert. If you have no specific vacancy yet, you can pick a typical advert for your profession. From those it writes eight questions for your panel, with no double-barrelled questions and no follow-ups packed in advance. Swiss specifics are named in its instructions, including salary expectations, the Arbeitszeugnis (the Swiss employer reference) and notice periods.

Reason 2: you speak instead of reading

An answer you have composed in your head nearly always sounds different when you say it: longer, with an opening that hunts for its point and an ending that never arrives. That is why the tool is spoken rather than typed. A voice asks the question, you answer into the microphone, and a countdown shows that you have up to three minutes.

If an answer is vague, the voice may ask one short follow-up in the role of whoever asked, then wait. It does not score or coach you mid-interview. That recreates what counts on the day: an unfamiliar question, one attempt at the answer, and someone who does not let an evasive sentence slide.

Reason 3: the panel matches your invitation

In Switzerland you rarely face one person. Often it is your future manager plus someone from HR, and in larger firms a specialist or a team member too. The tool lets you set up to five roles and name them yourself, and the questions are shared out accordingly. With more than one person on the panel, the voice says who is asking.

You also choose the company size in four bands, from one to five employees up to more than two hundred. That changes the tone. In a micro business the German-speaking panel will use du, the informal you, and ask whether you really want to do a bit of everything; in a corporation it is Sie and structured answers in the STAR pattern. If you are moving from abroad, hearing both registers once before the real week of interviews helps. Practice runs in German, French, Italian or English.

Reason 4: the score is broken down and explained

A single number tells you little. Each answer therefore gets an overall score from 0 to 10 and three part scores: substance, whether there is evidence and a concrete example; structure, whether the answer has a shape a listener can follow; and fit, whether it suits this role and this panel. Two or three sentences explain the score, and one or two say what a strong answer would have added.

After the round, the panel reads all your answers together and writes a debrief: a summary, strengths, concerns, a note from each role, and a percentage for how likely this round would have been passed with this panel. We read that number cautiously and so should you. It measures your answers against a rubric; it knows nothing about the other candidates or the mood in the room. It is a practice figure, not a forecast of an offer.

Reason 5: the free limit is stated plainly

Without an account, two questions are generated and the first answer is scored. A free account adds five scored answers on the day you sign up and five a month after that. The subscription is CHF 15 a month or CHF 95 a year, starts with a seven-day trial and covers every tool on the site. The subscription has a ceiling too: 40 scored answers a month, 80 on the yearly plan, because spoken minutes are the most expensive part of the tool to run.

We print those numbers because they are the part vendors like to leave out. Knowing them, you can decide whether the free account covers a single interview, which it often does, or whether a subscription pays off for a week with several appointments.

When we recommend something else

If what unsettles you is the other person, the eye contact, the silence, the handshake, a voice on a laptop will not train that. A friend, a colleague or a paid interview coach who can see you is the better choice. The same goes when presence and body language are part of the decision, as in a senior role or in sales: the scoring reads a transcript of your answer and sees neither posture nor eye contact.

If you are registered with a Swiss RAV (the regional employment office), ask about an application course with interview practice first; it costs you nothing but takes time to arrange. If you only want relaxed conversation practice in English with no job attached, the free voice mode of a general chatbot is enough. And if your doubt is not how to speak but whether the role or the career move is right, you need career guidance, not rehearsal.

There are technical limits as well. With a strong accent, a dialect or a poor microphone the transcript can mishear words, and then the tool scores an answer you did not quite give. Always read the reasoning before you believe a low score.

Test our recommendation in ten minutes

Take a real advert you intend to apply for and your current CV. Read the two questions the tool writes and ask yourself whether a hiring manager would put them that way. Answer the first aloud, without notes. Then read the reasoning and the note on the strong answer, and check whether you can turn it into one specific change.

If the questions feel generic or the reasoning tells you nothing new, the tool is not for you and you have paid nothing to find out. If both hold up, you know from your own experience why we recommend it, and you no longer have to take our word for it.