Build Everything Like You'll Have to Delete It
Your code will be wrong. Not might be—will be.
The question isn't whether you'll make bad decisions. It's whether those decisions will be survivable.
The Fork in the Road
Every line of code is a bet on the future. What if you switch providers? What if the API goes down for a week? What if you need to support something you can't even imagine yet?
If the answer is "we'd have to rewrite everything," you've planted a landmine.
The road ahead is full of forks. Your code should be indifferent to which path you take.
What Crypto Taught Me
At my last startup, I spent six months on three separate projects that regulators later forced us to shut down.
After the third one, something clicked.
I started building everything with the assumption it might get deleted. Database flags to toggle entire features. Abstractions so clean you could repurpose something by changing a dropdown.
Here's the weird part: when you build for reversibility, you end up following "best practices" by accident. Not because someone told you to—because it's the only way to stay sane.
The Real Skill
Anyone can build something that works today.
The real skill is building something that survives contact with tomorrow.
