Inspiration
I wanted to bridge the gap between theory and real IT interviews. Preparing for technical interviews can be stressful, and candidates often struggle to get human-like, personalized feedback. The idea of an AI-powered interviewer came from observing how repetitive practice and immediate evaluation can drastically improve coding and communication skills.
What it does
A real-time AI-powered interview platform that:
- Acts like a human interviewer, asking relevant technical questions.
- Provides instant feedback, scoring answers, and suggesting improvements.
- Tracks conversation history to tailor follow-up questions.
How I built it
Node.js + Express for the API server, handling real-time requests. React.js for the UI/UX. OpenAI API for evaluating answers.
Challenges ran into
OpenAI API sometimes returned 429 Too Many Requests. I implemented a fallback default question to ensure the app remained usable.
What I learned
Prompt engineering is crucial: crafting clear instructions for AI drastically improves question relevance. Error handling in real-time systems matters; fallback defaults keep the user experience smooth.
What's next for AI Interview Simulator
Add more IT topics (system design, database design, cloud architecture)... and also helping users improve soft skills.
Log in or sign up for Devpost to join the conversation.