Inspiration
The inspiration for AUTOCODE.IDE came from a simple but powerful insight: while coding is one of the most valuable skills in today’s world, less than 0.5% of the global population can actually code. Many beginners and even experienced developers struggle with syntax, logic, debugging, and adapting to new technologies such as AI. We wanted to change that. We built a platform that not only helps users write better code but also empowers those with minimal experience to explore advanced AI concepts—like building their own neural networks—without needing to master every technical detail. Our vision is to make coding and AI creation accessible to everyone, regardless of background or experience.
What It Does
AUTOCODE.IDE is a smart, web-based coding environment enhanced by AI and equipped with tools to guide users in real time. It provides:
AI Code Assistant
Powered by OpenAI API, the IDE can help you write, fix, and improve your code across multiple programming languages. Get instant suggestions, explanations, and debugging help.
IDE with Real-Time Preview
A sleek, customizable code editor with live output display, built for productivity and learning.
Code Translator
Easily convert code from one language to another (e.g., C++ to Python), helping learners understand language differences and logic translation.
Neural Network Builder
An interactive visual tool that allows users to build, configure, and experiment with basic neural networks—Currently still a alpha version prototype, could be improved in the future.
Project Workspace
Users can organize their work, save projects to github, and run small-scale projects directly within the browser.
How We Built It
Back when we were in grade 8 and 9 from informatics class, our teacher asked us to make html,js, and python code using trinket. We found using trinket is hard because it always has an error, also not accurate form what it shows and what it is supposed to show. so this web helps coders fix, improve, and see the output of their code without needing to run and debug it like vs code which saves a lot of time.
🔹 Backend (Optional/Extensible)
Currently no backend is required for basic code execution, keeping it lightweight and fast.
Future enhancements may integrate Node.js and cloud storage for user data and advanced computation.
🔹 AI Integration
OpenAI API – Provides code completion, debugging help, and natural language interaction.
Challenges we ran into
The challenging part of this hackathon is when we're finding a project suitable for the theme. We use the first 40 minutes only for finding the suitable project. We also run into a lot more problems such as how to make real-time updates without a backend. Implementing real-time code previews and translations all entirely on the frontend was technically complex. Without a backend, performance optimization and handling state changes smoothly became key concerns.
What we learn
This is our first offline hackathon, we learn valuable lessons about time management, collaboration, and teamwork. During this 30 hours of work we worked together on one whole website. We are very happy to be given the opportunity to participate in Garuda Hacks 6.0. We hope that our idea can help most coders, especially those who were just starting to try coding.
What next for autocode IDE
- multiple files integration, because right now it can only work on one file
- improve neural network builder, AI is growing rapidly and not all people can follow the growing speed, with this project we hope it could help.
Built With
- chatgpt
- codemirror
- css
- html
- javascript
- netlify
- openai
- pyodide
Log in or sign up for Devpost to join the conversation.