I'd question your premises -- AI makes the early boilerplate easy. Some people can clone the surface level UX of a SaaS in a week. Technical skill is absolutely needed to push beyond those two points. And technical implementation skills to code a SaaS has always been table stakes, not a competitive moat.

If anyone can clone most SaaS products in a week, then I already am pretty impressed. But assuming it's true, I would also assume that "anyone" is somewhat exclusive to people who already understand or have the ability to understand how existing SaaS products work. So that just means every company should have a team of people creating whatever SaaS products they need that week. Except that would create so much unreasonable amount of codes to support that the point would be that coding would simply be a race to build edge case scenarios to account for everything. which is also a huge issue. and so i think that the true skill to have that will offer competitive advantage is people management. alternatively some domain knowledge. since domain knowledge will give you the standpoint to decide what you should be doing. but i think a lot of this is kind of overblown. like you aren't really going to want an AI generated on the fly payments system like Stripe or an HR system like tripletex. You are going to want one that simply works and solves your problem and not something your team hacked together in a week with AI.

Technical skill has never been valuable after a certain point except for rare exceptions.

If you look at the leveling guidelines of any major tech company. Technical skill alone only gets you to a mid level job. After that it’s about scope, impact and dealing with ambiguity.

https://www.levels.fyi/blog/swe-level-framework.html

https://dropbox.tech/culture/sharing-our-engineering-career-...

A competitive advantage is knowing how to communicate business value from using technology.

> Anyone can clone most SaaS products in a week

I give you the challenge to clone Jira or GitLab in a week. I'll give you a $million if you can do either.

Intentionally avoiding AI and making that a core value of the product.