🛠 What it does
BugBot is an AI-powered Python debugging assistant. You give it a natural language prompt, and it:
Generates clean Python code
Asks for any required input parameters
Executes the code in real time
Detects and fixes errors on the fly using an LLM
Installs any missing modules automatically Basically, it’s like having a helpful Python TA sitting next to you 24/7.
⚙️ How we built it
Frontend/UI: Built entirely in Streamlit, styled with custom CSS for a clean and modern feel.
AI Backend: Uses Groq API to power code generation and self-healing debugging logic.
Execution Engine: Python's built-in exec() and ast parsing were used to safely run and modify generated code.
Input Parsing: Extracts function parameters using regex + AST and prompts users dynamically.
Deployment: Render (free tier) for simple, fast cloud deployment.
😵💫 Challenges we ran into
Making the code repair loop stable—AI doesn’t always fix bugs in one go.
Handling unknown or missing Python modules at runtime.
Keeping the UX minimal while allowing advanced debugging.
Parsing code dynamically without breaking or overfitting to certain function styles.
🏆 Accomplishments that we're proud of
Fully functional debugging loop using natural language input 🎉
Auto module detection + installation (saves SO much time)
A clean, responsive frontend built with zero JS
Solo-developed and deployed in under a week
🧠 What we learned
How to chain LLM calls effectively for interactive tools
Handling Python code execution and error repair safely
Dealing with deployment bugs under pressure
Building a minimal product that’s actually useful and not just flashy
🔮 What's next for BugBot
Add support for multiple programming languages (JavaScript is next!)
Let users upload files and debug entire scripts
Add dark mode + syntax highlighting
Package it as a VSCode extension
Release it as a desktop app via Electron or PyInstaller
Log in or sign up for Devpost to join the conversation.