Intervu

• 

  Shows results of the interview and your grade based on your responses.

• 

  live feedback in categories based on your response to the question.

• 

  shows the interview process where a question is asked and the user has to respond to the question.

What inspired me: Interviews are stressful, and most people do not get enough practice before the real thing. I wanted to create a tool that feels like a safe space to rehearse, make mistakes, and improve without needing a friend or career coach on standby.

What I learned: I explored how AI can simulate natural conversations, generate dynamic follow-up questions, and deliver constructive feedback. I also learned a great deal about balancing user experience with technical performance, and how small design tweaks can make the platform feel more human.

How I built it: I built Intervu using Cursor, which helped me streamline coding and iterate quickly. I created a clean, minimalist interface and used AI models for natural language generation to simulate realistic interviews and provide instant feedback. I worked on the frontend with React and Tailwind, then connected backend APIs to handle the question flow, scoring, and response analysis.

Challenges I faced: The biggest challenge was making the AI feel natural instead of robotic. Early versions sounded stiff and repetitive, so I refined the prompts and added variation to improve realism. Another challenge was designing feedback that was actionable. Finding the right balance between encouragement and critique took several iterations.

Built With

• css
• cursor
• git
• html
• javascript
• npm
• openai-api
• react
• tailwind-css