InterviewGenie: AI Talent Suite

Index or home page
dashboard
Resume Scanner
Report analysis

💡 Inspiration The recruitment process is often broken for both sides. Candidates struggle with generic feedback and "black hole" resume systems, while recruiters are overwhelmed by hundreds of applications. We built InterviewGenie to humanize AI in recruitment—using Gemini 3 Flash to provide a bridge between a piece of paper (the resume) and a real conversation.

🚀 What it does InterviewGenie is an end-to-end AI talent ecosystem:

For Candidates: It provides a futuristic Resume Scanner that calculates ATS compatibility and a Live AI Mock Interview room with real-time voice interaction and webcam integration.

For Recruiters: It features a Bulk Ranking Engine powered by FAISS to instantly rank hundreds of candidates against specific job descriptions.

🛠️ How we built it We utilized a robust Python-based stack:

Backend: Flask handles the orchestration between our different engines.

AI Core: Gemini 3 Flash powers the semantic analysis of resumes and the intelligent dialogue in the interview room.

Vector Intelligence: We used sentence-transformers and FAISS (Facebook AI Similarity Search) to build our ranking engine for high-speed document retrieval.

Frontend: A high-performance UI built with Tailwind CSS, featuring interactive Chart.js skill maps and real-time audio visualizers.

🧠 Challenges we faced Integrating real-time speech recognition with AI latency was a hurdle. We optimized the prompt engineering for Gemini 3 Flash to ensure that the "Live Session" feels like a natural conversation rather than a robotic exchange.

🏆 Accomplishments that we're proud of Successfully implementing a bulk ranking system that can process massive datasets in seconds and creating a truly immersive "Interview Room" that uses the user's camera and voice to simulate a real-world pressure environment.

📈 What's next for InterviewGenie We plan to add Body Language Analysis using computer vision to give candidates feedback on their non-verbal cues and eye contact during the mock sessions.