Skip to content
InterviewLah
Start practising

Proof of concept · staging preview

Practise job interviews with an AI interviewer.

Answer real interview questions out loud and get instant feedback on what you say, how you say it and how you come across, so you walk into the real thing ready.

Start a practice interview Instructor dashboard

No sign-up needed for this preview · works best in desktop Chrome or Edge

Ms Tan, Store Manager: an AI interviewer with an AI-generated face of a fictional personAI interviewer
Ms Tan · Store Manager“A customer is upset because an item they wanted is out of stock. What would you do?”

Instant feedback · sample

  • Eye contact78%
  • Speaking pace128 wpm
  • Filler words3
  • Answer structureGood

2D avatar mode · Mr Arjun

course areas
8
course areas
interview scenarios
12
interview scenarios
interviewer personas
12
interviewer personas
for Brightspace
LTI 1.3
for Brightspace

How it works

From nervous to ready in four steps

Each practice session takes about 5–8 minutes. Practise as often as you like.

  1. Step 1: Pick a role

    Choose your course area and a job scenario, from retail associate to junior web developer.

  2. Step 2: Talk to the AI interviewer

    Answer out loud on camera. The interviewer listens and asks follow-up questions based on what you said.

  3. Step 3: Get instant feedback

    See a rubric-based report on your content, delivery and presence, with clear tips for next time.

  4. Step 4: Track your progress

    Practise again and watch your scores improve. Instructors see trends for each learner and class.

Features

Feedback on the whole interview, not just the words

Content, delivery and presence each count in a real interview. The coach looks at all three and explains what to work on.

  • An interviewer that adapts

    Every scenario has a role, a workplace and an interviewer persona. Follow-up questions respond to your actual answers, like a real interview.

  • How you say it

    Speaking pace, filler words (“um”, “like”, “actually”), long pauses and answer length, measured from your speech with word-level timings.

  • How you come across

    Eye contact, facial expression, head steadiness and posture, analysed on your own device. Your video is never uploaded.

  • Rubric-based feedback

    Instructors set the criteria, weights and level descriptors. Learners get a clear report after every session with strengths and next steps.

  • Dashboards for instructors

    Individual, class and course views with trends and criterion breakdowns. Export to CSV or Excel, and connect to Brightspace via LTI 1.3.

  • Interviewer personas

    A lightweight 2D avatar by default, which runs entirely in the browser, plus an optional photoreal video interviewer as a premium mode.

Meet the interviewers

An interviewer for every industry

Each scenario comes with its own interviewer: their role, tone and the questions they care about. All interviewers are AI. 8 of them also have a photoreal face; those faces are AI-generated images of fictional people.

Courses & scenarios

12 scenarios across 8 course areas

Scenarios are written for real entry-level roles in Singapore workplaces. Instructors can edit them or add their own, with their own rubrics.

For instructors

See where every learner needs help

Set up scenarios and rubrics, then follow progress by learner, class or course. Spot common weak points and plan your next lesson around them.

Open the instructor dashboard

  • Scenario & rubric editor

    Roles, questions, interviewer persona, criteria, weights and level descriptors.

  • Class & course dashboards

    Trends over time and a breakdown for every rubric criterion.

  • Brightspace via LTI 1.3

    Designed to launch from a Brightspace course, with roster sync and grade return.

  • CSV & Excel export

    Download learner and session results for your own records.

Privacy & trust

Private by design

We collect as little as possible. The less we keep, the less can go wrong.

  • Video stays on your device

    Your camera feed is analysed in your browser. Video is never uploaded or stored.

  • Only results are kept

    We keep your answer transcript and summary scores, such as an eye-contact percentage. No images, no face data.

  • Not used to train AI

    Your answers are not used to train AI models, ours or any provider's.

  • AI processing in Singapore

    Speech-to-text, the interviewer's voice and the AI interviewer run on Google Cloud in Singapore (asia-southeast1).

About the photoreal interviewer: this optional premium mode uses a specialist avatar provider whose servers are outside Singapore. It receives only the interviewer's own lines (its voice or text), never your voice, video or personal details. The default 2D interviewer runs entirely in your browser. Optional live captions in Chrome are off by default and are explained in the notice.

Read the full privacy & AI notice

FAQ

Questions, answered

Which browsers work best?

Use the latest desktop Chrome or Microsoft Edge for the best experience. Other modern browsers work too; if voice answers are not available, you can type your answers instead.

Do I need a camera and microphone?

They give you the full experience: your browser will ask for permission when a session starts. If you say no, you can still practise with typed answers, and the camera-based feedback is simply skipped.

Is my video recorded or uploaded?

No. Your camera feed is analysed inside your browser and only summary numbers, such as an eye-contact percentage, are kept. You can replay your answers during the session; the recording is discarded when you leave the page.

Who are the interviewers?

They are AI characters, not real people. Their photoreal faces are AI-generated images of fictional people. The main questions come from scenarios that instructors set up; the follow-up questions are generated by AI from your answers.

How long does a practice session take?

About 5 to 8 minutes: up to five questions plus follow-ups, then your feedback report.

Can it connect to our learning platform?

It is designed to plug into Brightspace through LTI 1.3: learners launch it from their course, and rosters and grades flow back. Results can also be exported as CSV or Excel.

Is this the finished product?

No. This site is a proof of concept on a staging server. All learners, classes and results shown in the app are synthetic.

Your next interview starts with practice

Pick a role and talk to an AI interviewer now. It takes about five minutes.

Start a practice interview Instructor dashboard