Releasing a framework isn’t about writing the framework.
That part is easy.
Boyyee...the real work is everything around it:
docs, guides, examples, a website, launch prep, Product Hunt, tweets, answering questions before they’re asked - and I even created a dedicated InfrontJS GPT.
So this weekend I locked myself in for ~48 hours and finished all the non-code work. No new features. Just making the framework real.
Wait? Why just another frontend framework??
Frontend frameworks have become moving targets.
Breaking changes, endless abstractions, constant churn.
InfrontJS goes the other way:
- minimal core
- explicit APIs
- no compiler tricks
- browser-first
- boring by design
If it works today, it should work next year.
Stability over novelty
InfrontJS is intentionally boring.
That’s a feature.
The core is small enough to understand in one sitting.
Once something lands, it stays.
Frameworks should behave like infrastructure, not fashion.
Funny side effect: AI loves it
While building the InfrontJS GPT, it became obvious:
AI tools are really good with this framework.
- Small surface area.
- Predictable structure.
- No hidden lifecycle magic.
Turns out models love boring code too.
What happens now
Nothing exciting.
No feature requests.
No roadmap theater.
Just bug fixes, real-world usage, and time doing its job.
To be honest with you - this release was actually planned last year.
But like many side projects, it sat in the backlog while “more urgent” things happened. Since I’m starting the year with a hard backlog cleanup, I decided to stop postponing it, lock myself in for a weekend, get the shit done — and ship.
Ahh ... and if you are still reading this - you might be interested in a link to the project:
Thanks for reading.
The backlog is clean. New projects can begin.
Top comments (0)