What an AI mock interview is good for
An AI mock interview gives you an interviewer whenever you have twenty minutes. It asks questions, listens to your spoken answers, sometimes follows up, and then gives feedback. It won't replace a friend who knows your industry, but it is good at three jobs: getting you used to answering out loud, showing you where answers are vague or long, and letting you repeat a weak answer until it improves.
It works best as rehearsal. Treat it like a sports practice: short, frequent, focused on one thing at a time.
Set it up like the real thing
The closer practice is to the real interview, the more it carries over. Give the mock interview the job description and your résumé if it accepts them, so the questions match the role. Sit where you'll sit on the day, with the same device. Answer out loud at full voice; typing or whispering trains a different skill.
If your real interview is a recorded one, set a timer to the real limit. If it's a live screen, practice with follow-up questions turned on. Our post on practicing follow-up questions explains why those matter.
During the session
Don't stop to restart every time you stumble. Real interviews don't let you, and recovering mid-answer is part of what you're practicing. Say 'let me put that another way' and carry on.
Keep each answer to the length the real interview will want: about a minute for 'tell me about yourself', up to two minutes for a story. The earlier post in this series on answer length and pauses has targets for each kind of question.
Reading the feedback
Feedback is only useful if it leads to a change. When you read it, look for the one or two answers that were weakest, not a list of everything. Then ask why each one was weak. The usual reasons are few:
- It didn't answer the question that was asked.
- It had no specific example.
- It was unclear what you did yourself.
- There was no result.
- It ran long before the point arrived.
Pick the reason that fits and fix just that. If the feedback suggests a better version, borrow its shape, but put it in your own words with your own details.
Redo the weak answer, then the whole interview
Straight after reading the feedback, answer the weak question again out loud, two or three times. Each round should be a little tighter. Then leave it.
A day or two later, run a full mock interview again. If the same weakness shows up, work on that answer on its own a few more times. If it doesn't, move to the next one. Over a week, this usually takes an answer from rough to something you can say calmly.
I'm good at working under pressure. In my last job things were often busy, and I always got things done.
In my last job we lost two people from a team of six in the same month. I took over their weekly supplier orders, wrote a checklist so nothing was missed, and kept every order on time until we hired again.
How to tell it's working
You'll notice three things. Your answers start shorter and get to the point sooner. You stop needing notes for your main stories. And the first question stops feeling like a shock. When your answers sound like you talking rather than reciting, you're ready. Stop there; over-rehearsing makes answers stiff.
Hmmlo is one way to do this: Maya, an AI recruiter, or Nora, an AI hiring manager, asks about your own résumé, and Cal, an AI coach, reviews each interview and shows how a better answer can sound. You can see a sample feedback report to know what to expect.
