Inspiration
As a career coach, I saw countless talented professionals stumble in interviews due to lack of practice. I wanted to create a tool that makes high-quality interview preparation accessible to everyone—no expensive coaching needed. The goal was to replicate real interview scenarios with AI, so users build confidence and skill through realistic practice.
What it does
InterviewSimulator is an AI-powered platform for interview practice across industries/roles. Users customize simulations (type/difficulty/company style), respond to AI-generated questions, and get instant feedback on their answers (scored across 5 key dimensions). The dashboard tracks progress, and a library of 500+ questions with expert frameworks helps users master responses.
How we built it
We built the frontend with React for a responsive UI, paired with Node.js/Express for the backend. MongoDB stores user data/interview history; AWS hosts the platform for scalability. OpenAI API powers question generation/follow-ups, while Stripe handles premium subscriptions. Chart.js visualizes performance metrics, and Firebase manages user auth.
Challenges we ran into
Fine-tuning the AI to generate contextually relevant follow-up questions was tough—we iterated on prompt engineering to ensure questions matched user responses. Balancing scoring consistency across dimensions required rigorous testing, and enforcing non-premium usage limits (3 interviews/month) without breaking UX took careful state management.
Accomplishments that we're proud of
We created a scoring system that’s both objective and actionable, with feedback that users say feels personalized (not generic). The dashboard’s improvement trends have helped users track progress tangibly, and the Best Answer Library has become a go-to resource for job seekers across 20+ industries.
What we learned
We learned the importance of user-centric iteration—early testers highlighted the need for more industry-specific question banks, so we expanded the library by 300%. We also discovered that visualizing progress (vs. just showing scores) drastically boosted user engagement with the platform.
What's next for this project
We’re adding voice-based interview simulations (using speech-to-text APIs) to mimic real verbal interviews. We’ll expand the question library to cover 1000+ roles, add industry-specific readiness assessments, and integrate with job boards to align practice with active applications. We also plan to launch a mobile-responsive redesign for on-the-go practice.
Log in or sign up for Devpost to join the conversation.