Intervue_Bloom

• 

  Intervue_Bloom Home page

🚀 About the Project — Intervue_Bloom 🌱 Inspiration

Intervue_Bloom was inspired by a common problem faced by many students and freshers: fear of interviews and the lack of meaningful feedback after practice. Traditional mock interviews often feel unrealistic, and candidates rarely know what exactly went wrong. I wanted to build a system where users could practice interviews in a safe environment and receive clear, actionable feedback to improve with confidence.

🛠️ How I Built the Project

Intervue_Bloom is a full-stack AI-powered mock interview platform built using Next.js for the frontend and Tailwind CSS for a modern, responsive UI. User authentication and data management are handled using Firebase Authentication and Firebase services. The core interview experience is powered by Gemini API for intelligent question generation and response evaluation, while Vapi enables real-time, voice-based interview interactions. Interviews are dynamically generated based on the user’s selected role and technology stack, making each session personalized and realistic.

🤖 Key Features

Voice-based AI mock interviews

Dynamic question generation based on role and skills

Real-time evaluation of responses

Detailed feedback on content, communication, and clarity

Performance summaries and improvement suggestions

⚙️ Challenges Faced

One of the major challenges was handling voice latency and accuracy during live interviews. Integrating and synchronizing the Vapi workflow with AI response evaluation required careful state management and optimization. Ensuring smooth real-time interaction while maintaining accurate feedback was technically demanding but rewarding.

📚 What I Learned

Through this project, I gained strong experience in full-stack architecture, especially integrating AI into real-world applications. I learned how to design AI-driven workflows, manage authentication and data using Firebase, and build scalable, interactive systems. Most importantly, this project helped me understand how to combine AI, voice interfaces, and full-stack development to solve a real user problem.

Built With

• firebase
• firestore
• geminiapi
• nextjs
• tailwind
• typescript
• vapiai