All code will translate to tech debt sooner or later. The more code you write, the more bogged down in tech debt you will find yourself in the future.

Therefore, when starting a new project, you must ruthlessly prioritize building stuff that you're highly confident will help you make money. Much of the time, you will be wrong anyway, and you'll accumulate unproductive tech even if you try not to.

The worst thing you can do when you're starting a business is letting yourself invest in features that require many lines of code and have zero chance of making the company more money.

My contention is that if you do this as an entrepreneur, you will almost certainly kill your business.

If you want to get paid well as a software engineer, you should at least recognize when you're working on a product that will go nowhere. Don't work for a loser.