Forem: Shayan Ijaz

The Part of Programmig AI still can't touch....

Shayan Ijaz — Mon, 19 Jan 2026 07:06:23 +0000

The Noise Around AI Replacing Programmers

There’s a lot of noise claiming that AI will completely replace programmers and make this skill worthless. Sounds crazy, right?

Before panicking, let’s take a step back and look at history.

Lessons from Calculators and Mathematicians

Remember when calculators came into mathematics? They were faster and more accurate at calculations than humans.

Did that mean mathematicians disappeared? No!

Calculators didn’t exist to replace mathematicians—they existed to make calculations easier. This freed humans to focus on logic, reasoning, and problem-solving, while the calculators handled the repetitive work.

AI in Programming ... the same principle

AI in programming works similarly.

It can write boilerplate code.
It might even generate an entire codebase.

But managing the project, making decisions, and understanding the logic behind the code remains the programmer’s job. AI is a tool, not a replacement.

Why we still need to Understand the Fundamentals?

Does this mean we should stop learning how to code? Not at all.

It’s similar to arithmetic: if a calculator can do the basic operations but you don’t understand them, you’re blindly trusting a tool. You need knowledge first, then leverage the tool.

AI works the same way. Without understanding programming fundamentals, you’ll just blindly accept what it produces.

My Personal Experience Using AI

even used AI to generate boilerplate code while building my first terminal-based library in C.

It sped up the process significantly. All I had to do was understand the infrastructure and then let AI handle the repetitive parts—just like a calculator helping with arithmetic.

This allowed me to focus on the thinking and structure behind the code, rather than typing everything from scratch.

i am not good in articulating my thoughts...even i used AI to help me write this blog in a formal way...(im working on it!!...)

Takeaway

My take? Use AI as a tool to bring your vision to life faster, not to replace your thinking.

Learn, understand, and then let AI help accelerate the parts that don’t require your judgment. That’s where real programming value lies—and what AI still can’t touch.

Pixel-Powered Encryption: Turning an Image into a Key with Python

Shayan Ijaz — Fri, 02 Jan 2026 17:59:56 +0000

> Ever wondered if an image could be your secret key? Not a password, not a string of numbers—an actual picture. Each pixel contributes to your encryption key, and without the exact same image, your message is unreadable.

That’s exactly what I explored in a Python prototype, and today I’ll show you how it works and how you can try it yourself.

Why an Image as a Key?

Images are basically data. Each pixel has red, green, and blue values (RGB). If we take these values and combine them in a predictable way, they can serve as a key stream for encryption.
PPM format (P3) is perfect for this experiment: it’s a plain-text image format, easy to parse.
Each pixel = three numbers (R, G, B) between 0 and 255.
The order of pixels matters—mess with one pixel, and decryption fails.
Think of it like this: your message is a lock, and your image is the key. Only the exact key image can open it.

How the Encryption Works

We use a simple XOR encryption:

Convert your message to bytes.
Convert the image pixels into a keystream. For each pixel, we collapse RGB into a single byte: 3.

key_byte = (R + G + B) % 256

XOR each message byte with the corresponding key byte.
Decryption is the same process—XOR is symmetric. Note: This is not secure encryption for real secrets. It’s a concept experiment for learning and fun.

Why It’s Interesting

Conceptual simplicity: You don’t need advanced crypto knowledge to see how it works.
Visual intuition: The key is literally an image.
Experiment-friendly: You can modify pixels, shuffle, layer multiple images, etc.
Even though it’s not secure, it’s a great playground for:

Learning about image file formats (PPM)
Understanding basic encryption concepts
Exploring creative coding ideas

Try it or Contribute on : https://github.com/urshayan/PpmCrypt

Ruby GUIs Are Dead… Or Are They?

Shayan Ijaz — Thu, 25 Dec 2025 22:25:23 +0000

So I wanted to make a simple desktop app in Ruby — just a GitHub stats fetcher. You know, input a username, get their avatar, followers, repos… easy stuff.

Except it wasn’t that easy.

What’s up with Ruby GUI gems?

Tk: basically dead on modern Ruby

Shoes: fun idea, totally abandoned … Even the website seems to be crazy when downloading the executable

Scarpe / FXRuby / Qt: either unstable, heavy, or impossible to set up

What worked

I ended up using GTK3. It works. Cross-platform …and was easy to setup…

What I learned…

Ruby is amazing for scripting, APIs, automation — but GUI support is neglected.
Always check if a gem is maintained before committing time.
Community contributions matter — a little patch here or a fork there can keep gems alive.

Why it matters

If you’re a Ruby dev, or learning Ruby, don’t get discouraged. The tools are messy, but the language is still a joy. And if enough people care …maybe the GUI ecosystem can come back alive.

Nonetheless, GTK3 works. Other gems mostly don’t. But hacking through it is part of the fun