Inspiration

Writing and maintaining code can be time-consuming. Developers often spend significant effort understanding existing code, identifying improvements, and making it cleaner and easier to maintain. We wanted to build a tool that makes this process faster and more accessible.

What it does

Refactor.ai is an AI-powered code refactoring assistant. Developers can enter code and receive an improved version along with explanations of the changes.

Key features include:

AI-powered code refactoring
Explanations of suggested improvements
Code insights to help developers understand changes
Side-by-side comparison of original and refactored code

How we built it

We built Refactor.ai using React and Vite for the frontend and Python for the backend. The application uses the Groq API with the GPT-OSS-20B model to generate refactored code and explanations.

We deployed the frontend on Vercel and the backend on Render.

Challenges we faced

One of our main challenges was connecting the frontend and backend reliably in a production environment. We worked through deployment configuration, API routing, and environment variables to get the complete workflow running.

Another challenge was making the results easy to understand. We focused on presenting the original code, improved code, and explanations in a clear interface.

What we learned

We gained practical experience building and deploying a full-stack AI application, integrating an LLM API, handling production configuration, and designing an interface around AI-generated code.

What's next

We would like to expand language support, improve refactoring accuracy, add more customization options, and explore additional models and inference platforms.