[geophile]

The article gets at this briefly and moves on: "I can do all of this with the experience on my back of having laid the bricks, spread the mortar, cut and sewn for twenty years. If I don’t like something, I can go in, understand it and fix it as I please, instructing once and for all my setup to do what I want next time."

I think this dynamic applies to any use of AI, or indeed, any form of outsourcing. You can outsource a task effectively if you understand the complete task and its implementation very deeply. But if you don't, then you don't know if what you are getting back is correct, maintainable, scalable.

[SoftTalker]

> instructing once and for all my setup to do what I want next time.

This works up to a point, but eventually your "setup" gets complicated, some of your demands conflict, or have different priorities, and you're relying on the AI to sort it out the way you expect.

[tsycho]

But setups get equally complicated, even with human software engineers. The approach that the OP is talking about applies only to experienced, good architect-level SWEs, and I suspect that the code quality and its problems are going to be the same whether they are directing LLMs vs a set of junior SWEs to write the code.

There is an inherent level of complexity in projects that solve some real world problem, due to all the code handling edge cases that were added incrementally over time.

[eqvinox]

> any use of AI, or indeed, any form of outsourcing

Oh that's a good analogy/categorization, I hadn't thought about it in those terms yet. AI is just the next cheaper thing down from the current southeast asian sweatshop labor.

(And you generally get what you pay for.)

[bodge5000]

On the face of it, this or at least acting as a code reviewer from an experienced point of view seems like the solution, the problem is that we all naturally get lazy and complacent. I actually think AI was at its best for coding a year or so ago, when it could kind of do part of the work but theres no way you could ever ship it. Code that works today but breaks in 6 months is far more insidious.

[zmmmmm]

It does beg, the question , whether any of this applies to less experienced people. I have a hunch that the open-ended nature of what can be achieved with AI will actually lead right back to needing frameworks, just as much as we do now, if not more, when it comes to less experienced people.