Mock interviewBeta

An AI mock interview built from the interview you have next

Tsenta's AI mock interview is a practice session with an AI interviewer, built from your upcoming interview's company, role and round, your résumé, your saved stories and the job description when Tsenta has one. You speak your answers by default, and typing is a choice; the interviewer pushes on vague answers, and when you end, a report shows scores, STAR feedback and what to work on next.

Mock interview is in the Interview tab of the Tsenta dashboard on the web, tagged Beta. The agent that applies for you also reads your interviews from the job email you connect, so each one shows up in My interviews, ready to practice. It covers recruiter screens, hiring manager interviews, technical rounds and panels in 20 role families, not only tech.

Start practicingIn Beta, in the Tsenta dashboard on the web.
The Interview tab's mock interview card for a Google technical round, with Speaking selected, Tailor and a 6-question plan.
The plan for a Google technical round: 6 questions in about 40 minutes, answered out loud.
2,000+
mock interview questions, in role-family packs and a general pack
20
role families with their own question pack, tech and non-tech
4
round types: recruiter screen, hiring manager, technical, and panel or onsite
72
companies whose published values the interviewer knows
How it works

How to practice for an interview with AI on Tsenta

  1. Step 1

    Tsenta finds your interview

    Tsenta reads upcoming interviews from the job email you connect, or you add one by hand with the company, role, round and a day or time. Each interview gets a card in My interviews with a Mock interview button, and Next up at the top names the one thing to do first, such as Practice it now.

  2. Step 2

    Check the plan, or tailor it

    Before you start, the session card shows the plan under This session covers: the topics and competencies, your focus areas, the minutes and the interviewer. Tailor lets you change up to 3 focus areas, the length, the interviewer and the level.

  3. Step 3

    Speak your answers

    Speaking is already picked on the session card, so press Start interview and talk. Your words appear in the thread as you say them. Switch to Typing if you would rather write. Questions come one at a time, and the interviewer pushes when an answer leaves out what you did or what came of it.

  4. Step 4

    Read your report

    Shortly after you end, the report scores each answer, marks the STAR parts each behavioral answer gave or missed and names what to work on next. Retry this question practices any one question again, or you can start a mock built around the thing to fix.

Built from your real interview

A mock interview based on the job description, your résumé and your stories

The mock for an interview is built from its company, role and round, your résumé and your saved stories. When the interview is linked to one of your applications, it also uses the job description Tsenta has for the posting. For a Google technical round, the session card's plan covers your background, learning (marked as a focus), incidents and on-call, system design, languages and frameworks, and project planning and estimates. In that example it also notes a gap: "You have no story yet for ambiguity or learning."

For 72 companies, the interviewer also has the company's published values. It does not invent facts about the company.

Apart from the company's published values, the mock reads only your own data. Spoken audio is not stored by Tsenta, and your answers and the report are stored in your account.

See everything in Interview prep

Four round types

Recruiter screen, hiring manager, technical and panel interview practice

A mock started from an interview card uses that interview's round. The round sets how many questions you get and about how long the mock runs.

For people in tech roles, Coding and System design sit beside Mock interview in the same tab, for the parts of a technical loop you practice in an editor or on a whiteboard. A technical round's card gets a Coding questions button only when the company is in Tsenta's coding index. The role must also be in software engineering or data science, or Tsenta must count you as someone in a tech role.

Recruiter screen (phone screen)
5 questions, about 25 minutes.
Hiring manager interview
6 questions, about 40 minutes.
Technical interview
6 questions, about 40 minutes.
Panel, onsite or final round
7 questions, about 45 minutes, asked by 2 or 3 fictional panelists.
Behavioral
Pick a round yourself and the hiring manager asks mostly for stories.
Short version
3 questions in 15 minutes, for any round.

See coding interview practice

Speaking by default

AI interview practice out loud, with a live transcript as you talk

A mock is practice for talking, so speaking is how it starts. The interviewer's questions are read aloud, you answer out loud and press Done when you are finished. Your words appear in the thread while you speak, so you can see what the interviewer heard before you hand the answer over. Typing is one tap away on the session card if you would rather write.

The interviewer says it is an AI practice interviewer and asks one question at a time. Mock interview is practice before the real thing, and Tsenta does not join, listen to or record your real interviews.

Live transcript
The words you say show up as you say them, then the answer goes to the interviewer when you press Done.
Typing
Pick Typing on the session card, or switch mid-interview; the mock runs the same either way.
No audio kept
Tsenta keeps the words, not the recording.
A harder interviewer

An AI interviewer that pushes back instead of nodding along

The interviewer talks like a person, not a script. It opens in its own words, calls you by your first name now and then, and moves on with a short bridge rather than a summary of what you just said. It never praises you, leads you or hints whether you would pass.

A follow-up is one question, usually a sentence or two, anchored in your own words. It digs when an answer is missing your own action, a result or a number, so "we shipped it and it went well" gets a question about what you did or what came of it. Wave a question away and it does not move on: it takes the claim at face value in a few words, then makes the question harder with a real constraint. After two vague replies it moves on.

Supportive
For entry level rounds: it follows up on gaps patiently, with no curveballs.
Demanding
The usual setting: it pushes on vague or confident claims, asks for numbers and the option you rejected, and plays devil's advocate once.
Tough
For senior rounds: curveballs on most questions, and it asks for failure modes as a matter of course.

See feedback on a real interview

STAR feedback

Behavioral interview practice with STAR method feedback

On each behavioral answer, the report checks situation, task, action and result, plus a number in the result. Each part shows as given, given after a follow-up, or missing, with a tip for each part you missed. A summary line counts how many behavioral answers had every STAR part without a follow-up, such as 2 of 3.

Every answer is scored on structure, specifics, relevance and length, and the report sets how you answered next to a stronger answer. Retry this question runs that one question again, so you can fix a missing task without sitting through a whole mock.

Tsenta's report on one mock answer, with a stronger answer, four scores, STAR chips marking the task missing, and Retry this question.
This answer left out the task, so the Task chip is marked missing and a Task tip appears.
Your report and your stories

Mock interview feedback: what to work on next and a STAR story bank

The report opens with the short version of how it went and shows job fit against the job description when Tsenta has one. What to work on next names the first thing to fix, with a mock built around it.

Your stories is a STAR story bank you edit: a title, the situation, task, action and result, and the competencies each story shows. Answers you gave in a mock can be kept as stories, and the report names the story you used and one of yours that fits the question better. The interviewer knows your story titles and favors questions your stories can answer.

After the real interview, add its transcript for feedback on what you actually said. Then Prep for the next round gives you a mock of your next round, built around that feedback.

See feedback on a real interview

Tsenta's report on a Stripe hiring manager mock, with the short version, four scores and what to work on next.
The first fix for this Stripe mock: say what you were responsible for in each story.
Not only tech

Mock interview practice for non-tech roles

The question bank holds 2,000+ questions across 20 role-family packs and a general pack. The families are software engineering, data science, product management, design, IT operations, sales, marketing, finance, consulting, operations, human resources, legal, healthcare, education, research, and mechanical, electrical, civil, chemical and industrial engineering.

So a mock for a teaching role draws on the education pack, and a mock for an accountant draws on the finance pack. Mock interview and Feedback are there for every role. Tsenta shows Coding and System design to people in tech roles, based on the roles you search for and apply to, your interviews and your résumé.

AI mock interview questions, answered

An AI mock interview is a practice round where an AI interviewer asks you one question at a time, and in Tsenta it is built from an interview you have coming up. You speak your answers, with your words showing in the thread as you talk, or switch to typing, and the interviewer follows up when an answer is missing your own action, a result or a number. Shortly after you end, Tsenta shows a report with per-question scores, STAR feedback and what to work on next.

Yes, Tsenta builds each mock interview from the company, role and round of your interview, your résumé and your saved stories. The mock also uses the job description when the interview is linked to one of your applications and Tsenta has the job description for the posting. Tsenta reads upcoming interviews from the job email you connect, and an interview you add by hand has no job description unless you link it to an application. For 72 companies the interviewer also has the company's published values, and it does not invent facts about the company.

For hiring manager interview prep or panel interview practice, run a Tsenta AI mock interview for that exact round. A hiring manager mock is 6 questions in about 40 minutes, and a panel, onsite or final round mock is 7 questions in about 45 minutes with 2 or 3 fictional panelists. Recruiter screens are 5 questions in about 25 minutes, technical rounds are 6 in about 40, and a short version of any round is 3 questions in 15 minutes. Before you start, Tailor changes the focus areas, length, interviewer and level.

Yes, and speaking is how a Tsenta mock interview starts, because a mock is practice for talking. The interviewer's questions are read aloud, you answer out loud, and your words appear in the thread while you speak, then you press Done. Typing is a choice on the same card if you would rather write. Spoken audio is not stored by Tsenta.

Tsenta's interviewer is built to push rather than nod along. It never praises you or hints whether you would pass, it asks one question at a time in its own words, and it digs when an answer is missing your own action, a result or a number. Brush a question off and it raises the bar with a concrete constraint instead of moving on. How hard it pushes follows the level: patient at entry level, pushing on vague or confident claims by default, and curveballs on most questions at senior level. Tailor changes the level before you start.

No, a Tsenta mock interview score rates a practice answer and does not predict a hiring decision. Each answer is scored on structure, specifics, relevance and length so you can see what to fix, and the AI interviewer never hints whether you would pass. The report also names what to work on next, the first thing to change before the real interview.

You can practice the STAR method by answering behavioral questions and checking which parts you gave: situation, task, action, result and a number in the result. In a Tsenta mock interview, the report marks each part as given, given after a follow-up, or missing, with a tip for each part you missed. Retry this question lets you answer the same question again, and Your stories keeps a STAR story bank you can edit.

Yes, Tsenta's mock interview has 2,000+ questions across 20 role-family packs and a general pack, so a nursing mock draws on the healthcare pack and a sales mock on the sales pack. The families also include marketing, finance, consulting, operations, human resources, legal, education and research, as well as engineering, design and product roles. Tsenta shows Coding and System design to people in tech roles.

No, Tsenta is not a live interview assistant, and it does not join, listen to or record your interviews. Tsenta is built for practice before an interview and feedback after it. Its mock interviewer says it is an AI practice interviewer. After a real interview, you can add the transcript to Tsenta for feedback and a mock of the next round.

Your mock interview answers and the report are stored in your Tsenta account, and spoken audio is not stored by Tsenta. Apart from the company's published values, the mock reads only your own data, such as your interview, your résumé, your stories and the job description Tsenta has for the posting.

Mock interview

Practice your next round before the real one

Start practicing

Tsenta finds the interview in the job email you connect, or you add it, then builds a mock for that company, role and round and shows you what to fix first.

Markdown for AI agents