

Built a realtime AI Interview Assistant prototype that listens to interview questions and generates short natural answers live
Features:
• Live transcript
• Realtime AI answers
• Resume + JD personalization
• Adjustable answer length
• Sensitivity control
• Auto-clear old answers
• Users use their own Gemini API key (stored locally in browser/device only)
No accounts.
No backend AI costs.
No voice output.
Just realtime assistance.
Currently testing latency + realtime streaming performance.
Would you actually use something like this for interview prep/practice?
Open to feedback and feature ideas.