InterviewIQ

Inspiration

Preparing for technical interviews is often inefficient and frustrating. Most students rely on platforms like LeetCode for practice, but they don’t receive structured feedback on their solutions. On the other hand, general AI tools provide answers, but not a consistent system for tracking progress or identifying weaknesses. We wanted to build a tool that simulates a real interview preparation experience — where users can practice, receive meaningful feedback, and improve over time.

What it does

InterviewIQ is an AI-powered interview preparation platform that helps users practice coding interviews in a structured way. Generates company-specific interview questions based on difficulty and type Allows users to submit their solutions in multiple programming languages Provides structured AI feedback: Score and correctness Time and space complexity Strengths and weaknesses Missed edge cases and optimizations Tracks user progress over time Displays readiness level and weak topics through a dashboard

How we built it

Frontend: React-based interface with pages for problem generation, coding workspace, and progress dashboard Backend: Flask API with modular routes for: Problem generation Solution feedback Progress tracking AI Integration = Groq LLM used to generate interview questions and structured feedback in JSON format. Database = Stores generated problems and user attempts for persistence and analytics Architecture: Designed as a full-stack system where frontend interacts with backend APIs, and all user activity is stored and analyzed

Challenges we ran into

Getting consistent, structured AI responses required careful prompt design and strict JSON formatting Handling edge cases where AI responses were malformed or incomplete Designing a system that feels like an “interview experience” rather than just another chatbot Managing state between frontend, backend, and database during rapid development Time constraints in implementing both generation, evaluation, and analytics features