Oh no, the opposite. So many people will be writing software via claude now. So many more software-based, or at least, software-integrated businesses are going to pop up. And when as they grow, LLMs aren't going to be enough.
What's going to change is that we'll have to focus on the actual hard problems of software development: making sure the software does what it's supposed to, isolating nontrivial bugs, finding ways to make it perform better.
> actual hard problems of software development: making sure the software does what it's supposed to, isolating nontrivial bugs, finding ways to make it perform better.
These are exactly the things LLM's outperform all of us on.
I don't understand the vision of the one-person Claude-powered business. If it can be done with one person, it can be done with zero more efficiently. What seems more likely to me is the emergence of a few (or one) massive vertically-integrated everything firms, who can afford more inference than anyone else seeking to enter the market. What we call AI, is more than that, it's accummulation, capital, systems of power, and these juggernauts will simply have more of it. Agents will continually do research to seek out areas of market inefficiency and octopus into them. They can outline business requirements, spec them out, and delegate to other agents. Other agents can continually research perf improvements, hunt for vulns, and so on. Once the race to the bottom kicks off, there's no undoing it. It seemed absurd to me a few months ago but every time I've thought capabilities were limited in some fundamental way I've been proven wrong