DEV Community: Arshvir

I create my own AI OS

Arshvir — Mon, 06 Apr 2026 17:08:00 +0000

Hi guys, my name is Arshvir and today I’m documenting my journey to build an AI Operating System. I named it as “TURING AI OS” built on top of Linux. You will be surprised that I tested it on my fourteen-year-old PC and the name is inspired from the famous computer scientist Mr. Alan Turing. A huge respect for him. Now back on topic so, I heavily customized KDE Neon which itself is built on top of ubuntu and ubuntu is based on Linux Kernel.

So, What Exactly is Turing AI OS? What feature it have? and what about our privacy?

But before answering these question, I gonna tell you that I have face so many problems, bugs, errors and lot of things during the development of this OS.

Initially I choose Linux Mint as base OS but later on I ditched it, then I choose pure debian then again I ditched it. After that I finally choose KDE Neon as base OS because it is light in version and have minimal bloatware.

Turing AI OS is fully functional ai os which actually thinks and works for you. It has features like:

AI Sidebar (Your Personal Private Copilot)
Right Click File/Folder AI Analyzer
AI Control Panel (for managing memory and models)
Mini AI Spotlight
AI Terminal (aka Turing Shell)
Customized Theme
Bundled Dev and Privacy Tools

I use Ollama as a Backend AI engine, langchain for LLM interaction and agentic work, Qwen3 billion parameter model as default but you can use any which is supported by ollama and PyQT6 for frontend.

You can run this OS with minimal requirements as follow:

Minimum i3–3rd Gen
Minimum 14teen year old PC.
Electricity (Optional)

Below are useful links for more information:

Thanks :)

Radiant AI: Building an AI-Powered Image Editor with Python

Arshvir — Thu, 01 Jan 2026 18:26:41 +0000

Hi Developers, This side Arshvir :)
Today, I deployed an Advanced AI Image Editor and Processing Tool on Hugging Face spaces. You can Explore the app by clicking on the here: Radinat AI

So, you might curious what is actually Radiant AI? Well Radiant AI is an Advance Image Editor and Processing Tool powered by AI and several Deep Learning Models. The main idea behind Radiant AI is to combine traditional image editing with modern AI-based computer vision in one application.

what can it do?
It includes various key features such as:

Restoring your old or degraded photos
Colorize the black & white or old photos
Upscaling images so, there will be no compromise with quality
Enhance image quality
Turn your photos into sketch, cartoon style and you can even use various filters.
Apply Creative Filters such as Sepia, CyberPunk, HDR and etc. Alongside, it also provides basic editing and utility tools, including:
Brightness, contrast, saturation, sharpness
Flip or rotate your image
Background removal
Privacy blur, object/person blur
Format converter, resize image and batch processing.

Technologies I Used
I used a variety of technologies to build Radiant AI:

Backend: Python, Flask
Frontend: HTML, CSS, JavaScript
Deep Learning Models: YOLOv8, EDSR (Super Resolution), and others
Libraries: OpenCV, rembg, YOLO
Deployment: Hugging Face Spaces
Planned: Offline desktop application

I'm currently exploring more optimized deep learning model to make Radiant AI best of the best. Furthermore, I'm also working to make it as desktop application and turn this application into Micro SaaS App.

Radiant AI is still an early project, but it’s been a great learning experience.
If you have any feedback, ideas, or feature suggestions, I’d love to hear them.

Below are my GitHub and LinkedIn Profiles:

Thanks for reading!

I built a privacy-first IP API and it's free.

Arshvir — Wed, 17 Dec 2025 16:23:18 +0000

Hello developers, This side Arshvir from Punjab, India. I just built a privacy-first IP tools and API and it's free. It includes various feature for both developers and general users. When you visit jaiho ip then it shows your current IP address, Location, ISP, OS information, and Browser information.
and for developer's you guys can use this as API also which include all features as listed on website.
Below is code snippet that how to use What's My IP? Jaiho-IP

For Python Developer:

import requests

response = requests.get('https://jaiho-ip.vercel.app/api/json')
data = response.json()

print(f"IP Address: {data['ip']}")
print(f"Location: {data['location']['city']}, {data['location']['country']}")
print(f"ISP: {data['isp']['name']}")
print(f"OS: {data['client']['os']}")

For JavaScript:

fetch('https://jaiho-ip.vercel.app/api/json')
  .then(response => response.json())
  .then(data => {
    console.log('My IP:', data.ip);
    console.log('Location:', data.location.city + ', ' + data.location.country);
    console.log('ISP:', data.isp.name);
    console.log('Browser:', data.client.browser);
  })
  .catch(error => console.error('Error:', error));

cURL: curl -s https://jaiho-ip.vercel.app/api/json