Inspiration Preparing for technical interviews is often inefficient and scattered. Most candidates struggle with finding high-quality, role-specific questions, practicing consistently, and reviewing ideal answers. I wanted to solve this by building a focused, AI-powered platform that generates personalized interview questions and answers instantly — helping users prepare smarter, not harder.
What it does The AI Interview App allows users to enter any topic or job role (like “Machine Learning” or “Java Backend”), and instantly receive high-quality interview questions with model-generated answers. It leverages large language models via OpenRouter and supports user interaction through a clean, responsive interface. It also uses Supabase to store and retrieve question history for authenticated users.
How we built it Frontend: Built using React.js with Vite for fast performance and deployed on Vercel. It communicates with the backend using the native Fetch API.
Backend: Developed using FastAPI, exposing a single main endpoint (/agent-chat) to interact with OpenRouter AI models. Deployed on Render.
Database: Used Supabase for user authentication and storing question-answer history.
API Integration: Integrated OpenRouter to fetch AI-generated questions and answers using models like Gemma and DeepSeek.
Environment Management: Used .env files and configured them securely across Render and Vercel.
Challenges we ran into Integrating OpenRouter initially caused 400 errors due to incorrect headers and model names.
CORS setup between frontend (Vercel) and backend (Render) took time to configure properly.
Supabase integration required learning new auth and database query patterns.
Environment variables were tricky to manage consistently across local and cloud environments.
Accomplishments that we're proud of Successfully deployed a full-stack AI-powered app without Docker or Kubernetes.
Achieved real-time question-answer generation using cutting-edge LLMs.
Cleanly integrated Supabase for user-based data persistence.
Seamless communication between frontend and backend across cloud platforms.
What we learned How to build and deploy a production-ready full-stack app using FastAPI and React.
Efficient use of Supabase for authentication and storage.
Best practices for managing environment variables and API security in cloud-hosted apps.
Debugging and resolving CORS, async requests, and deployment-level integration issues.
What's next for AI Interview App Add support for favourite questions so users can bookmark and revisit them later.
Implement voice support for hands-free mock interview practice.
Set up CI/CD pipelines for automated testing and deployment.
Built With
- css
- fastapi
- fetch
- github
- html
- javascript
- openrouter-api
- python
- react.js
- render
- supabase
- vercel
- vite
Log in or sign up for Devpost to join the conversation.