I Built a Break Reminder Because I Forgot How to Blink

"I'll just finish this one function..." — Me, 4 hours ago, now with dried-out eyes and a spine shaped like a question mark.

The Problem Nobody Talks About

Here's the thing about being a developer: we're really good at solving other people's problems. Building systems that scale. Optimizing queries. Debugging that one weird edge case at 2 AM.

But ask us to stand up and drink water? Apparently that requires a PhD in self-care we never obtained.

I realized I had a problem when I looked up from my monitor after what felt like "a few minutes" of coding, only to discover that:

It was dark outside (it was morning when I started)
My water bottle was still full
My eyes felt like sandpaper wrapped in regret

Why Existing Solutions Suck

"Just use an existing break reminder app," you say. I tried. Here's what I found:

The Bloated Ones: Apps that want to track my productivity, sync with 47 services, send me motivational quotes, and probably report my bathroom breaks to the cloud. I just want to be told to look away from my screen. That's it.

The Ignorable Ones: A tiny notification that politely suggests "maybe take a break?" Brother, I am in the ZONE. That notification is getting dismissed faster than my Slack messages.

The Aggressive Ones: Apps that lock your entire computer. Great, now I've lost my flow state AND my unsaved work. Thanks.

I wanted something different. Something that:

Actually interrupts you (because that's the point)
Doesn't feel like punishment
Gets out of your way when you're done
Doesn't require a 47-page privacy policy

Enter AFK

So I built AFK — a minimal break reminder that does exactly one thing well: it makes you step away from your keyboard.

Here's the philosophy:

1. Fullscreen, Because You Will Ignore Anything Less

When it's break time, AFK takes over your entire screen with a calm, minimal overlay. No, you can't just cmd+tab away from it. That's the point. Your eyes need a break, and your willpower clearly isn't cutting it.

2. Respects Your Flow

AFK detects when you're idle and pauses the timer automatically. Step away for coffee? The timer pauses. Come back? It resumes. No penalty for actually taking breaks on your own (shocking, I know).

3. Minimal by Design

No accounts. No sync. No analytics. No "premium tier with advanced break scheduling powered by AI." Just a timer and a fullscreen overlay. That's it. That's the app.

The Tech Stack (For the Nerds)

Building a native macOS app in 2026 is interesting. You have options:

Electron: Sure, if you want your break reminder to use more RAM than Chrome
Swift/AppKit: The "proper" way, but I wanted cross-platform eventually
Tauri: Rust backend + web frontend, native performance, tiny bundle size

I went with Tauri.

Why Tauri?

Bundle Size: The entire app is ~8MB. An equivalent Electron app would be 150MB+ (most of which is just Chromium sitting there, judging you).

Performance: Rust handles the system-level stuff (tray icon, timer, fullscreen overlay). React handles the UI. Best of both worlds.

Native Feel: It actually feels like a macOS app, not a web page pretending to be one.

The Tricky Bits

Fullscreen Overlay: Getting a window to cover the entire screen, including the menu bar, while still allowing keyboard shortcuts to dismiss it? Harder than it sounds. macOS has opinions about window levels.

// The magic incantation to create an unmissable overlay
let window = tauri::WindowBuilder::new(
    app,
    "break",
    tauri::WindowUrl::App("break.html".into())
)
.fullscreen(true)
.always_on_top(true)
.decorations(false)
.build()?;

Menu Bar App: AFK lives in your menu bar, not your dock. Click the icon, adjust settings, forget it exists until break time. This required some careful state management between the tray and the settings popover.

Idle Detection: Detecting when the user is idle (so we can pause the timer) uses macOS's CGEventSource. If you haven't touched your keyboard or mouse in a while, you're probably already taking a break.

The Indie Hacker Journey

Building AFK was a weekend project that turned into a month-long obsession. Here's what I learned:

Distribution is Harder Than Building

Writing the code? Fun. Figuring out how to:

Sign the app for macOS
Notarize it with Apple
Create a Homebrew cask
Set up a landing page
Handle the 47 different ways macOS can block your app from running

Not as fun.

Perfection is the Enemy of Shipping

The first version of AFK was ugly. The timer was off by a few seconds sometimes. The settings UI was basically unstyled HTML.

But it worked. And shipping something that works beats perfecting something that doesn't exist.

People Actually Want This

I shared AFK with a few developer friends. The response was immediate: "Wait, this is exactly what I needed." Turns out I'm not the only one who forgets to blink.

Try It Yourself

AFK is free. No catch. No "free tier with limitations." Just free.

Install via Homebrew:

brew tap Harry-kp/tap && brew install --cask afk

Or download directly: afk-app.vercel.app

What's Next?

A few things on the roadmap:

Windows/Linux support (Tauri makes this theoretically easy, but "theoretically" is doing a lot of heavy lifting)
Custom break screens (maybe you want to see a cat instead of a minimal gradient)
Break statistics (for those who want to track their eye-care journey)

But honestly? The app does what it needs to do. Sometimes the best feature is no new features.

The Real Lesson

Building AFK reminded me why I got into programming in the first place: to solve problems that annoy me personally.

Not every project needs to be a startup. Not every app needs to scale to millions of users. Sometimes you just need to build something that makes your own life slightly less painful.

And maybe, just maybe, reminds you to drink some water.

Now if you'll excuse me, AFK is telling me it's break time. And for once, I'm actually going to listen.

Built with sleep deprivation and spite by Harry-kp and Chaitanya