InterviewPrep

The Inspiration✨

We want to leverage AI in solving real-life problems and our group all had anxiety when dealing with interviews and then it hit us. What if we could make an application that helps train and give feedback when practicing for an online interview?

What it does

It provides scores for confidence, anxiety, engagement by analyzing facial expressions and speech clarity to help prepare students for their job interviews. It then gives tailor tips on what areas to improve during an online interview call based on those scores.

How we built it 🔨

We built InterviewPrep using a full-stack approach with modern technologies:

Frontend

Designed the user interface in Figma for a clean, intuitive experience
Built with React and JavaScript for dynamic interactions
Styled using Tailwind CSS and HTML for responsive design

Backend & AI

Developed RESTful APIs using FastAPI for efficient request handling
Integrated AWS Rekognition for facial recognition and emotion detection
Utilized AWS Transcribe for accurate audio transcription
Processed interview data with Gemini AI to generate comprehensive scores for confidence, emotional state, and engagement levels

Collaboration & Version Control

Managed code and collaboration through GitHub

Challenges we ran into 💡

Since this was our first time using AWS for facial recognition and audio transcription, we ran into some errors and had to do debugging to solve the issues. During implementation, we encountered issues with our initial Google API integration for facial recognition and audio transcription, which led us to pivot to the DeepSeek API. Debugging was further complicated by logs not appearing in our console, making it difficult to trace errors. Through researching API documentation and testing different configurations, we successfully resolved these issues and got the scoring system working.

Accomplishments that we're proud of 🎓

We are proud to create and submit a full stack project leveraging AI to assist students prepare for their interviews. Learning new frameworks and tools together under time pressure. Designing a clean, functional website that is both technical and engaging. Building a meaningful project inspired by our personal experiences.

What we learned📚

Today, we gained valuable experience in integrating multiple technologies to build a cohesive AI-driven system.

Team Collaboration is about listening as much as contributing
User Experience matters as much as technical excellence
FastAPI how to handle backend requests and connect with AWS services for facial recognition and speech-to-text transcription.
Multimodal Data analyze and process it combining video and audio inputs to extract meaningful insights like confidence, emotion, and engagement levels.

What's next for InterviewPrep 🌟

We have a lot of implementation ideas to expand InterviewPrep's capabilities:

Real-time Feedback - Implement live analysis during practice sessions to provide instant coaching
Personalized Questions - Tailor AI-generated questions based on specific job roles and industries to create more relevant practice experiences
Enhanced Interview Preparation - Build a comprehensive system that adapts to each user's career goals and target positions