DEV Community: Palak Verma

GhostScript: A Haunted Terminal That Brings the Machine to Life

Palak Verma — Wed, 28 Oct 2026 14:52:00 +0000

GhostScript: A Haunted Terminal That Brings the Machine to Life

🕸️ What I Built

For the Ghost in the Machine challenge, I created GhostScript, a spooky, interactive terminal inspired by Halloween and the mysterious world of computers.

GhostScript gives users the feeling of communicating with a haunted machine. It combines a dark terminal interface, glowing effects, and interactive commands to create a fun horror-themed experience.

Features

Haunted Terminal: Interact with a mysterious computer interface.

Ghost Story Generator: Discover random horror stories.

Dark Jokes: Enjoy spooky programming jokes.

Summon a Ghost: Receive mysterious messages.

Interactive Commands: Type commands such as help, story, haunt, and joke.

Spooky Design: Enjoy a dark purple theme, glowing text, and floating ghost animations.

Responsive Interface: Works on desktop and mobile screens.

Technologies Used

I built this project using simple web technologies:

HTML5 — for the structure of the website.

CSS3 — for styling, animations, and responsive design.

JavaScript — for interactive commands and dynamic responses.
https://mahiiverma351-gif.github.io/GhostScript/
How It Works

Users can open the website in their browser and interact with the terminal.

For example:

Type story to generate a random horror story.

Type haunt to summon a mysterious ghost message.

Type joke to receive a spooky joke.

Type help to see all available commands.

The current version uses predefined JavaScript responses, making it lightweight, beginner-friendly, and usable without an internet connection.

Why Open Innovation Matters

I believe open innovation gives developers the freedom to experiment, learn, and build creative projects without depending entirely on closed platforms.

GhostScript is currently a browser-based prototype. My next goal is to integrate an open-weight language model through a local runtime such as Ollama.

This could allow GhostScript to generate original horror stories, respond more naturally to user prompts, and run locally after the model has been downloaded.

I am especially interested in exploring how local AI can offer more privacy, customization, and control to users.

Challenges and Learning

While building GhostScript, I practiced connecting HTML elements with JavaScript, handling user input, creating dynamic messages, and designing a responsive interface.

This project helped me understand how a simple idea can become an interactive web experience using basic coding skills.

What's Next?

I plan to improve GhostScript by adding:

Real open-weight AI integration.

AI-generated horror stories.

Typing effects and spooky sound effects.

More ghost personalities and commands.

Local AI inference for offline conversations.

Final Thoughts

GhostScript started as a fun Halloween-inspired coding idea, but it also gave me an opportunity to explore the possibilities of open-source AI.

I am excited to continue improving it and learning more about web development and local AI.

Build with curiosity. Code with creativity. Never trust a computer at midnight!

What would you add to a haunted terminal?

GhostInTheMachine #OpenSource #AI #WebDevelopment #JavaScript #Hacktoberfest #HalloweenCoding

SideHustle Studio: Build Better Product Listings with Open-Source AI

Palak Verma — Thu, 15 Oct 2026 10:33:39 +0000

SideHustle Studio: An AI-Powered Toolkit for Small Businesses 🚀

This is a submission for the Hacktoberfest Open-Source AI Challenge — Week 2: Side Hustle.

What I Built

I built SideHustle Studio, a web project designed to help small business owners and online sellers create better product listings and marketing content.

Many small sellers struggle to write attractive product descriptions, promote their products, and create content that catches customers' attention. They may not have the budget to hire a professional copywriter or marketing team.

SideHustle Studio aims to make these tasks easier by providing a simple, accessible tool for creating product content.

The project is designed for freelancers, independent sellers, and small businesses that want to save time and improve their online presence.

Demo

Live Website:

https://mahiiverma351-gif.github.io/SideHustle-Studio/

Code

GitHub Repository:

https://github.com/mahiiverma351-gif/SideHustle-Studio

Explore the project, review the source code, and follow its development on GitHub.

How I Built It

I built SideHustle Studio using front-end web technologies:

HTML to structure the website.
CSS to create the design and responsive layout.
JavaScript to add functionality and user interactions.
Open-source AI as the intended foundation for generating useful product content and helping sellers with repetitive marketing tasks.

My goal is to keep the project simple, practical, and easy to improve. The next step is to strengthen the integration with an open-weight model so users can generate content with greater control over their data and model choices.

Why Does Open Innovation Matter?

Open innovation matters because useful AI tools should be accessible to people who cannot afford expensive subscriptions or high API costs.

For a project like SideHustle Studio, open-weight models can offer several advantages:

Lower running costs: Local inference can reduce dependence on paid, per-request AI services.
Privacy: When inference runs locally, product information and business data can stay on the user's computer.
Freedom of choice: Developers can experiment with different open-weight models and select one that fits their needs.
Less vendor lock-in: The project can evolve without depending entirely on one company's pricing or platform.
Community-driven improvement: Other developers can inspect the code, suggest improvements, and contribute new features.

I believe open-source AI can help small businesses access useful technology while giving developers more freedom to experiment and build.

My First User

My next goal is to get SideHustle Studio into the hands of a small business owner or online seller, gather feedback, and improve the experience based on real needs.

I want to learn which features save sellers the most time, what they would improve, and whether they would pay for a tool like this.

Prize Categories

SideHustle Studio is submitted for the Side Hustle theme of the Hacktoberfest Open-Source AI Challenge.

Final Thoughts

Building this project has been an opportunity to explore how web development and open-source AI can help solve practical problems.

I want to keep learning, improve the project, and build tools that are genuinely useful to real people.

*Built with curiosity, creativity, and a belief in open innovation. *

I'd love to hear your feedback and suggestions for improving SideHustle Studio!

EcoQuest AI: Less Scrolling, More Wonder — Building an Open-Source AI Nature Companion for the Touch Grass Challenge

Palak Verma — Fri, 09 Oct 2026 12:06:05 +0000

# 🌿 EcoQuest AI — Less Scrolling, More Wonder

Live Website: https://mahiiverma351-gif.github.io/EcoQuest-AI/

GitHub Repository: https://github.com/mahiiverma351-gif/EcoQuest-AI.git

What I Built

EcoQuest AI is an AI-powered nature companion designed to encourage people to spend less time on screens and more time exploring the outdoors.

Built using HTML, CSS, and vanilla JavaScript, the application combines an AI-powered image scanner with outdoor challenges, a personal nature journal, activity statistics, and an outdoor break timer.

The goal is to use AI as a tool for real-world discovery rather than another reason to stay online.

Key Features

AI Nature Scanner: Upload a photo and receive image classification predictions from an open-weight AI model running in the browser.
Daily Outdoor Quests: Complete activities such as observing leaf shapes, listening to birdsong, and watching clouds.
Nature Journal: Save outdoor activities and AI-assisted discoveries locally in your browser.
Outdoor Break Timer: Take a ten-minute break and track your outdoor activity.
Activity Dashboard: View recorded outdoor minutes, saved discoveries, and activity streaks.
Responsive Design: Explore the application on desktop and mobile devices.

Tech Stack

HTML5
CSS3
Vanilla JavaScript
Hugging Face Transformers.js
MobileNetV4 image classification model
ONNX Runtime Web
Browser LocalStorage

How Open-Source AI Powers the Project

EcoQuest AI uses Transformers.js and a compatible open-weight image classification model to run AI inference directly in the browser.

Rather than sending every image to a proprietary AI API, the application processes selected photos on the user's device. This reduces dependence on remote inference services and helps keep personal images private.

The open model approach also makes it possible to experiment with alternative models and build more specialized nature-recognition features in the future.

The model and its required assets must be downloaded initially, so the current version is not yet fully offline.

Why I Built It

I wanted to explore how AI could encourage healthier digital habits and create a stronger connection with the natural world.

Many applications compete for our attention. EcoQuest AI takes a different approach: it gives users a small reason to go outside, notice their surroundings, and learn something new.

The idea behind the project is simple: technology should help us experience more of the world, not just spend more time looking at a screen.

What I Learned

Building EcoQuest AI gave me an opportunity to combine frontend development with browser-based AI inference.

I explored how to integrate a machine-learning model into a JavaScript application, manage user-uploaded images, save data locally, and create interactive features without requiring a dedicated application backend.

Future Improvements

Add specialized plant and bird identification models.
Improve offline functionality by caching model assets.
Add richer AI-powered explanations of nature discoveries.
Introduce location-aware nature trails and recommendations.
Improve journal storage and accessibility.
Test the experience during real outdoor activities.

Try It Out

🌐 Live demo: https://mahiiverma351-gif.github.io/EcoQuest-AI/

💻 Source code: https://github.com/mahiiverma351-gif/EcoQuest-AI.git

Take a quest, step outside, and discover something new.

EcoQuest AI — Built with curiosity and open-weight AI. 🌱