-
-
Home page
-
home page
Inspiration
AI coding tools can generate working code in seconds, but understanding that code is a different problem. Beginners often copy AI-generated code without knowing what each part does, why it works, or where something went wrong when it fails.
That inspired me to build ExplainMyAI — a simple tool that helps people understand AI-generated code instead of blindly trusting it.
What It Does
ExplainMyAI takes a piece of code and turns it into a clear, beginner-friendly explanation. It breaks the code down step by step, explains important concepts, highlights possible problems, and can simplify the explanation for someone who is still learning programming.
The goal is simple: don't just give people code — help them understand it.
How I Built It
I built ExplainMyAI as a focused proof of concept using:
- React + TypeScript + Vite for the frontend
- Python + FastAPI for the backend
- An LLM API for code analysis and explanations
- CSS for the interface
I kept the architecture intentionally small so the project could focus on one complete workflow: code goes in, the AI analyzes it, and the user receives a structured explanation.
I also used the Devpost Learn Skill Pack throughout the development process to plan the project before building it and to create the required scope.md, prd.md, and spec.md documents.
What I Learned
This project taught me that building with AI is not just about generating code. Clear planning, defining a small scope, structuring AI responses, and testing the complete user flow are just as important.
I also learned more about connecting a React frontend to a FastAPI backend, working with AI APIs, and designing an interface around information generated by an LLM.
Challenges
One of the biggest challenges was keeping the project focused. It was easy to imagine many additional features, but the hackathon required a small proof of concept that works end to end.
Another challenge was making AI-generated explanations useful rather than overwhelming. The interface needed to present technical information in a way that a beginner could actually follow.
That pushed me to focus on clarity, structured responses, and a simple user experience rather than adding unnecessary complexity.
What's Next
ExplainMyAI is currently a proof of concept. Future versions could include saved explanations, interactive code walkthroughs, more programming languages, personalized learning modes, and deeper debugging assistance.
For now, the core idea is simple:
AI can write the code. ExplainMyAI helps you understand it.
Log in or sign up for Devpost to join the conversation.