Reviving StylesByShahid: From an Unfinished Prototype to an AI-Powered Presentation Platform

Shahid Malik — Sat, 30 May 2026 09:39:41 +0000

This is a submission for the GitHub Finish-Up-A-Thon Challenge

This is a submission for the GitHub Finish-Up-A-Thon Challenge

What I Built

StylesByShahid is an AI-powered presentation creation platform designed to make building modern presentations faster, smarter, and more collaborative.

The project originally started as an experimental frontend prototype during my learning journey in full-stack development. Over time, it evolved into a larger vision involving AI-assisted presentation generation, authentication, real-time collaboration, and scalable backend architecture.

Using JavaScript, Node.js, Express, MongoDB, and GitHub Copilot, I revived the project and transformed it into a more complete and production-ready platform.

Key Features

AI-powered presentation workflow
Modern glassmorphism UI
JWT authentication system
Real-time collaboration using Socket.io
File uploads for multimedia presentations
Responsive design for desktop and mobile
MongoDB-powered backend architecture

Demo

GitHub Repository

https://github.com/SHAHID-glitch/StylesByShahid

Live Demo

https://shahidmalik786-stylesbyshahid.hf.space/#

Screenshots / Video Walkthrough

Before - the backend was incomplete and non-functional. User registration and login requests were not being processed correctly, making authentication unusable.

The application could not register new users, lacked database integration, and could not generate presentations using AI. Users were limited to static demo presentations.

Dashboard – The original version did not include a dashboard for managing presentations or user activity.

After - I fixed and completed the backend, enabling secure user registration, login, and authentication workflows.

AI Integration – I integrated Groq AI APIs to generate presentation content dynamically, transforming the platform from a static demo into an AI-assisted presentation builder.

Dashboard - I added a dedicated dashboard to improve navigation, presentation management, and the overall user experience.

Database Integration – I connected the platform to MongoDB, enabling persistent user accounts, presentation storage, and application data management.

GitHub Copilot in Action – GitHub Copilot assisted with backend development, API implementation, debugging, and code refactoring throughout the project revival process.

Video - https://player.mux.com/YMmB00CfqgOW2REobfWgAaNqNo4GSyzf01r3zG2gy00VSY

The Comeback Story

This project started as an unfinished idea that had strong potential but lacked polish, structure, and completion.

Originally, the platform only had:

basic frontend screens
incomplete backend integration
unfinished deployment setup
missing authentication flows
limited usability

For this challenge, I decided to revisit the project and properly finish what I started.

What I Improved

Refactored the project structure for better scalability
Improved backend API organization
Added deployment support with Docker and Hugging Face Spaces
Enhanced UI responsiveness and animations
Improved authentication and security handling
Cleaned up project architecture and documentation
Optimized the overall developer experience

The biggest transformation was shifting the project from an experimental prototype into a more realistic full-stack AI presentation platform.

My Experience with GitHub Copilot

GitHub Copilot played a major role throughout the revival process.

It helped me:

refactor repetitive backend logic
generate API boilerplate faster
debug integration issues
improve documentation
speed up frontend styling iterations
experiment with new implementation ideas

One of the most useful aspects was using Copilot as a collaborative coding assistant while rebuilding unfinished parts of the application.

Instead of spending hours searching for syntax or boilerplate patterns, I could focus more on product improvements and architecture decisions.

Tech Stack

Frontend: HTML, CSS, JavaScript
Backend: Node.js, Express.js
Database: MongoDB
Authentication: JWT
Real-time Communication: Socket.io
Deployment: Docker, Hugging Face Spaces

Project Impact

During this revival, I:

Completed the backend authentication system
Added MongoDB database integration
Integrated Groq AI for presentation generation
Added a dedicated dashboard
Improved deployment using Docker and Hugging Face Spaces
Refactored and organized the project structure

Final Thoughts

This challenge gave me the motivation to revisit an unfinished project and turn it into something much closer to the original vision I had in mind.

There’s still room to grow, but bringing this project back to life taught me an important lesson: unfinished projects still hold potential if you’re willing to return and improve them.

AI Error Doctor – A CLI That Explains Errors Like a Human

Shahid Malik — Wed, 04 Feb 2026 11:48:36 +0000

What I Built

I built AI Error Doctor, a simple but powerful Python CLI tool that takes a raw error message and instantly explains:

🔍 What the error means

🔧 Why it happened

✅ How to fix it

As a student and early-stage developer, I often found error messages confusing, especially for beginners. This tool is designed to act like a friendly debugging assistant right inside the terminal.

Example usage:

py error_doctor.py "ModuleNotFoundError: No module named 'requests'"

Example output:

📋 Error Type: ModuleNotFoundError
💡 Explanation: Python cannot find the requested module.
🔧 Suggested Fix: pip install requests

The tool currently supports common Python, npm, and Git errors, and it’s easy to extend with more patterns.

Demo
📸 Screenshots

💻 Code

https://github.com/SHAHID-glitch/ai-error-doctor

Video Link: https://player.mux.com/NY00L3RQlb6ZrZiPOT5pfl18WjfhnduhSMPqC7HsaaYw

GitHub Copilot CLI was central to this project.

I used it to:

🧠 Design the CLI structure

✍️ Generate error-handling logic

🔎 Suggest common error patterns and fixes

🚀 Rapidly iterate inside the terminal without switching contexts

Copilot CLI felt like pair programming in the terminal. Instead of searching error explanations online, I could build a tool that explains errors for others — using Copilot itself.

This project genuinely changed how I see debugging: from something frustrating into something automated, educational, and developer-friendly.

Final Thoughts

AI Error Doctor may be simple, but it solves a real problem faced by beginners every day.
This challenge helped me go from asking Copilot for help to building a tool that helps others.

Thank you, GitHub and DEV, for this opportunity 🙌

githubcopilot #cli #python #ai #devtools

DEV Community: Shahid Malik

Reviving StylesByShahid: From an Unfinished Prototype to an AI-Powered Presentation Platform

Project Impact

AI Error Doctor – A CLI That Explains Errors Like a Human

githubcopilot #cli #python #ai #devtools