AI Interview Question Generator

A lightweight AI-powered application that generates thoughtful, role-specific interview questions using the Gemini API.

Built with a strong focus on:

• simplicity
• fast iteration
• clean architecture
• maintainability
• product-oriented UX

---

Overview

This application allows users to enter a job title and instantly generate 3 tailored interview questions using AI.

The project was intentionally designed to prioritize:

• readable code
• lightweight architecture
• responsive user experience
• practical product decisions
• AI workflow clarity

---

Features

• Generate role-specific interview questions
• Gemini AI integration
• Loading states
• API error handling
• Responsive UI
• Clean typography
• Markdown rendering for AI responses
• Disabled submit state during requests

---

Tech Stack

Frontend

• React
• Vite
• Tailwind CSS

AI

• Gemini 2.5 Flash API

Deployment

• Vercel

---

Architecture

src/
├── components/
│   ├── JobForm.jsx
│   ├── QuestionList.jsx
│   ├── Loader.jsx
│   └── ErrorMessage.jsx
├── services/
│   └── gemini.js
├── App.jsx
├── main.jsx
└── index.css

Architectural Philosophy

The application intentionally uses:

• React hooks only
• minimal abstractions
• separated API layer
• lightweight component structure

This approach improves:

• readability
• maintainability
• iteration speed

---

Setup Instructions

1. Clone Repository

git clone <your-repository-url>

2. Navigate Into Project

cd ai-interview-generator

3. Install Dependencies

npm install

4. Configure Environment Variables

Create a .env file in the project root:

VITE_GEMINI_API_KEY=your_api_key_here

5. Run Development Server

npm run dev

---

Gemini Prompt Strategy

The application uses structured prompting to generate higher-quality interview questions.

Example:

You are an expert hiring manager.

Generate 3 thoughtful and role-specific interview questions
for a candidate applying for the role: "${jobTitle}"

Requirements:
- Questions should assess strategic thinking, communication, and role competency
- Avoid generic questions
- Return only a markdown bullet list.
- Do not include explanations or assessment labels.

The goal was to demonstrate:

• AI fluency
• prompt engineering awareness
• product-oriented output generation

---

Product Decisions

Why React + Vite?

• Fast development experience
• Lightweight setup
• Excellent iteration speed

Why Gemini Flash?

• Fast inference speed
• Strong free tier
• Simple developer experience
• Well-suited for lightweight AI workflows

Why Lightweight Architecture?

The project intentionally avoids:

• overengineering
• unnecessary libraries
• complex state management

Priority was placed on:

• clarity
• maintainability
• speed of execution

---

Error Handling

The application uses user-friendly error messaging:

Something went wrong. Please try again.

This avoids exposing raw API errors directly to users and reflects production-aware UX decisions.

---

Future Improvements

If given more time, potential enhancements include:

• response streaming
• prompt customization
• copy-to-clipboard functionality
• response history
• caching
• analytics
• rate limiting
• improved prompt consistency
• regeneration functionality

---

Deployment

The application is deployed on Vercel.

Add your deployment link here:

https://ai-interview-generator-two.vercel.app/

---

Loom Walkthrough

Add your Loom walkthrough link here:

hhttps://www.loom.com/share/3b63dd0d6a9b4369999258adfbd0f500

---

Philosophy

I believe software should be:

• understandable
• iterative
• user-focused
• practical

This project intentionally prioritizes clarity and product thinking over unnecessary complexity.