In this whole AI-written-code era, I've started feeling a little lost.
New AI tools, frameworks, projects, and software seem to be launching every hour. And somewhere in all of this, I started missing something very simple: writing code myself!
I miss brainstorming an idea, getting stuck on a problem, rummaging through piles of Stack Overflow tabs at 2 AM, trying things that made absolutely no sense, and finally figuring out what was wrong.
That feeling when something you've been struggling with for hours finally works? That dopamine hit.
So over the past few weeks, I've been looking for something I could get genuinely hooked on again. Something that would force me to sit down, think, experiment, break things, and actually write code.
And then an idea came up in my mind:
Why not build my own browser engine in Rust?
Not exactly a unique idea , there's already a real-life project, Servo, originally started by Mozilla. But as a frontend developer, building a browser engine from scratch felt crazy enough to be interesting. And I wanted to build it in Rust (why? just because I'm fascinated with the language).
There's just one small problem: I have a very basic understanding of Rust. So I'll be learning it along the way.
Why a Browser Engine?
I've always wondered what actually happens between writing HTML/CSS and seeing it appear on the screen.
- How does HTML become a DOM?
- How does CSS get parsed and applied?
- How does the browser figure out where everything goes, and eventually turn it into pixels?
I want to understand that by building a tiny version of it myself. Not because I think I can build a better browser , but because building one is probably a pretty good way to understand one.
AI Will Be Part of the Process
I'll absolutely be using AI. It'll help me understand concepts, learn Rust, break down problems, think through architecture, and get unstuck.
But I'll try to write as much of the actual code myself as possible.
The goal isn't the final codebase. The goal is the learning.
The Plan
My current milestones are (as helped by Claude to break it down for me):
- Project Skeleton
- Parse HTML → DOM Tree
- Parse CSS → Rules
- Style Resolution
- Block Layout
- Paint
- Text Rendering
- Basic Interactivity
I'm sure this list will change as I go.
The Real Goal
I'm not trying to build the next Chrome or Firefox. This is purely an educational project.
More than anything, I want to keep the coder in me alive. With AI making it incredibly easy to jump from an idea to working code, I don't want to lose the joy of figuring things out myself.
- The bugs.
- The bad code.
- The documentation rabbit holes.
- The late nights.
- And that moment when it finally works.
I'll be documenting each milestone, what I learn, and hopefully some of the things that go horribly wrong along the way.
Maybe this project goes nowhere. Maybe I abandon it halfway through. I'm okay with that , because the goal isn't really to build a browser.
Either way, I want to document the process.
Because with AI making it easier than ever to generate code, I think it’s worth occasionally slowing down and remembering what it feels like to actually write it.
Here’s to getting that coding dopamine back. 🚀
And yes, I might take a little help from ChatGPT to turn some of those learnings into articles. 😄
