Master System Design Interviews with AI-Powered Feedback.
SystemCraft is a high-fidelity system design interview simulator. It combines an interactive architectural canvas with a sophisticated AI engine to evaluate your designs against real-world constraints, providing deep insights into trade-offs, scalability, and structural integrity.
- Interactive Design Canvas: A powerful, intuitive workspace to build complex system architectures using industry-standard sub-components (LBs, Servers, Databases, etc.).
- Real-time AI Interviewer: Engage in simulated interview sessions where an AI evaluator monitors your progress and asks clarifying questions.
- Architectural Linter: Automatic detection of structural issues (e.g., disconnected load balancers, single points of failure).
- Deep Qualitative Evaluation: Powered by Google Gemini, the system analyzes your reasoning and provides a score (0-100) along with specific strengths, weaknesses, and suggestions.
- Session Management: Track your interview history and architectural improvements over time.
- Frontend: Next.js 15 (App Router), React 19, Tailwind CSS 4.
- Backend: Next.js API Routes (Serverless).
- Database: MongoDB Atlas with Mongoose ODM.
- Authentication: Firebase Auth (Google & GitHub).
- AI Engine: Google Gemini (via OpenRouter/Google AI SDK).
- Deployment: Vercel.
- Node.js 20+ installed.
- A MongoDB Atlas cluster.
- A Firebase project with Authentication enabled.
- An OpenRouter API key or Google AI Studio key.
git clone https://github.com/Shashank0701-byte/System-Craft
cd System-Craft
npm installCreate a .env file in the root directory (refer to .env.example):
MONGODB_URL=your_mongodb_connection_string
NEXT_PUBLIC_FIREBASE_API_KEY=your_key
# ... see .env.example for full listnpm run devPrefer containers? Check out our dedicated Docker Setup Guide for quick-start instructions using Docker Compose.
SystemCraft uses a hybrid evaluation strategy:
- Structural Analysis: Deterministic rules that check for physical connectivity and best practices in the canvas nodes.
- Reasoning Analysis: An LLM-based evaluator that examines the "Why" behind your choices, focusing on scale and trade-offs.
This project is licensed under the MIT License - see the LICENSE file for details.
Please create a feature branch before opening a Pull Request.