3.9 / 5

Interview Mirror

Live feedback on your video call presence and speaking

Useful

Real-time

Privacy

The Problem

Can't evaluate your own presence while in a conversation; feedback comes too late

Current Solutions (Not Great)

Record and review later (too late), expensive coaching, guess and hope

Who Needs This

Job seekers, salespeople, executives, anyone who does important video calls

Job interviews, sales calls, presentations—your on-camera presence matters but you can't see yourself objectively while talking. Interview Mirror runs during video calls and gives you real-time, non-distracting feedback: a subtle indicator when you're saying 'um' too much, when you've been looking away from camera, when your energy is dropping, when you're speaking too fast. It's like having a coach watching your performance and sending you tiny nudges without the other person knowing.

Honest Take

The coaching/interview prep market is hot and people pay for this. Competitors like Yoodli exist but there's room for differentiation. Technical challenge is doing all the analysis (filler words, eye tracking, energy detection) without killing the computer during an actual call.

Monetization Ideas

Ways to turn this into revenue

Freemium

Free basic, $5-15/mo for pro

Subscription

$5-29/month or $49-199/year

One-Time Purchase

$9-49 per license

Features

Key features that make this app valuable

Real-time filler word detection (um, uh, like)
Eye contact tracking (looking at camera vs. away)
Speaking pace monitoring
Energy/enthusiasm detection
Posture alerts
Subtle visual indicators (not distracting)
Post-call summary report
Improvement tracking over time

Build Prompt

Use this prompt with an AI assistant to start building

Build a React PWA called 'Interview Mirror' using WebInfer, MediaDevices, and Web Audio API. UI: small floating widget (draggable) with indicator lights, expandable post-call dashboard. On start: getUserMedia for video+audio (separate from whatever video call app is using). Video analysis (every 1s): face detection, eye gaze direction (looking at camera = good), posture check. Audio analysis (continuous): detect filler words via speech patterns, calculate speaking rate (WPM), analyze pitch variation (monotone = low energy). Use generateObject periodically to return { eyeContact: number, fillerWordCount: number, speakingPace: 'slow'|'good'|'fast', energyLevel: number, postureScore: number, currentIssue?: string }. Widget shows: green/yellow/red dots for each metric, subtle pulse animation when issue detected. NO audio or intrusive alerts during call. Post-call: full report with timeline, specific moments flagged, comparison to past calls. 'Practice mode' with more aggressive feedback for rehearsal.

Score Breakdown

Pain