Inspiration
Re.Code was inspired by a simple observation: most programming platforms focus heavily on teaching syntax, but almost none of them help developers practice debugging, one of the most critical real-world skills for hackers. During hackathons and while learning new languages, I often found that understanding why code breaks is far more valuable than memorizing how to write it. Re.Code was created to bridge this gap.
What it does
The platform generates realistic buggy code snippets and challenges users to fix them. Once a fix is submitted, the system evaluates the solution and explains what the bug was, why it occurred, and what best practices should be followed.
How we built it
I built Re.Code as a full-stack web application. The backend is developed using Flask. The core of the project is powered by the Gemini API, which is used both to generate realistic bugs and to evaluate user fixes with detailed, beginner-friendly explanations. The frontend is a lightweight HTML, CSS, and JavaScript interface. The application is deployed using Render.
Challenges we ran into
One of the main challenges was ensuring that the AI generated realistic and educational bugs rather than trivial or confusing ones. Balancing scope within a limited hackathon timeframe was also challenging
Log in or sign up for Devpost to join the conversation.