"does the test coverage adequately demonstrate the functionality of the code?"
and is this a solved problem? If not, then the bottleneck is right here, if it is solved, then yeah we shouldn't need anymore software engineers other than the elites
and is this a solved problem? If not, then the bottleneck is right here, if it is solved, then yeah we shouldn't need anymore software engineers other than the elites
yes. it was solved before but when writing code by hand the cost of building exhaustive test suites was far to high to do it in practice, except in very narrow cases where high assurance was required. now that AI can implement all of the testing frameworks for you it can be done for everything.
> we shouldn't need anymore software engineers
no. the job changes, but the skills that software engineers have are more valuable than ever because they now gate a much higher level of productive output.
corporations aren't really ruthless profit optimizers. micro incentives don't actually favor efficiency. hiring decisions don't actually have much to do with output and productivity. for example: it has been known forever that adding more people to a project usually decreases velocity, but that has never stopped anyone.
technology changes but people don't. AI makes higher quality software faster and at greater scale, and velocity is what is really valuable, so companies that master AI development will be making more money, and they will hire more people, because that is what they do.