80% of startups are going to die before their codebase reaches this size. They should probably not be making tech stack choices based on "what if we succeed beyond all likelihood".

two things come to mind: maybe those startups would die at a slower rate if their code was comprehensible to begin with ? but leaving that aside, for companies where survival is not an issue then, doesn't this indicate that using a dynamically typed language is not great?

I would be interested in seeing the number of startups that fail due to technical debt. My instinct is that most startups fail for business reasons (no clear need, not enough/right sales, poor management, bad pitch, solving the wrong problem, etc).

Well, in the case of Instagram, this just shows that Python was a great choice.

They went with a dynamically typed language, were successful, the language they chose added an optional type hinting system, and they wrote a tool that would automatically type hint their code in order to reap many of the benefits of a static type system.

I think that the amount of man hours that went into writing the tool is negligible, so it's a net win for Instagram.

New projects at large companies often are written in Java. New projects at startups aren't and shouldn't be - doubling future maintenance costs for the sake of a 20% reduction in development time now is a good trade, because 90% of startups fail, the important thing is to validate product/market fit as soon as possible.

(Though really you should just use Scala and get both Java-like safety and Python-like productivity)