InterviewIQ – AI Interview Simulator

• 

  This is frontend of my app in which user fill some details to start interview

• 

  This is second phase where user start their interview withAI

• 

  At thefinal stage, AI show the progress and final outcome of user

Inspiration

Interview preparation is often unstructured, stressful, and unrealistic. Most candidates rely on static question lists or generic advice that does not adapt to their role, experience level, or real interview performance.

The inspiration behind Interview IQ came from the idea that interview practice should feel like an actual interview where questions are asked one at a time, responses matter, and feedback is specific and actionable. I wanted to build a system that could simulate a professional interviewer and help candidates understand how they are performing, not just what they should study.

What it does

InterviewIQ is an AI-powered interview simulator that helps users practice technical interviews in a realistic environment.

Users select a job role, experience level, interview difficulty, and optionally a company. The system then conducts a structured interview by asking adaptive, role-specific questions one at a time. After the interview is completed, Interview IQ provides detailed feedback, including performance scores, strengths, weaknesses, and personalized improvement suggestions.

The platform focuses on realistic interview flow and meaningful evaluation rather than static questions.

How we built it

InterviewIQ is built using the Gemini API as the core intelligence of the application.

Gemini 3 Flash is used to conduct the live interview, acting as a professional interviewer. It manages question flow, adapts questions based on user responses, and maintains consistent interview pacing.

After the interview ends, Gemini 3 Pro analyzes the complete conversation transcript and generates structured feedback using JSON-based evaluation. This includes overall scores, category-based assessments, strengths, weaknesses, and actionable recommendations.

The frontend handles user interaction, interview progress tracking, and feedback visualization, while Gemini powers both conversational reasoning and analytical evaluation.

Challenges we ran into

One of the main challenges was making the interview feel realistic and controlled. Ensuring that the AI asked only one question at a time and followed a consistent interview structure required careful prompt engineering.

Another challenge was producing reliable, structured feedback instead of vague AI responses. This was addressed by leveraging Gemini’s structured output and reasoning capabilities.

Balancing simplicity for the user while maintaining advanced AI behavior was also a key challenge during development.

Accomplishments that we're proud of

Successfully using Gemini as the core logic of the application rather than a helper tool

Creating a realistic interview experience with adaptive questioning

Generating structured, actionable feedback instead of generic AI responses

Building a complete, end-to-end product suitable for real interview preparation

What we learned

This project taught us how powerful large language models can be when used beyond simple chat interfaces. We gained practical experience in prompt design, structured AI outputs, and designing AI-driven workflows.

We also learned how to translate AI capabilities into a product that solves a real-world problem in a clear and user-friendly way.

What's next for Interview IQ AI Interview Simulator

Future improvements include supporting more job roles, adding voice-based interviews, tracking progress across multiple sessions, and enabling resume-based interview personalization.

InterviewIQ can also be expanded for behavioral interviews, system design interviews, and enterprise-level hiring simulations.

Built With

• css
• frameworks:
• googleaistudio
• googlegeminiapi
• html
• javascript
• languages:
• nextjs
• platform
• react