I think “software is solved” splits into two levels, and the second one is much harder.
1. Local level: AI writes individual classes, functions, tests, and small features accurately and fast. No one’s writing code en masse by hand anymore.
← We’re basically here.
2. Macro level: shipping and running big real-world systems or full SaaS products that paying customers depend on 24/7.
The jump from 1 to 2 is brutal because 90% accurate isn’t good enough. At scale you need near-perfect reliability, observability, security, and the ability to evolve the system without it collapsing.
You also need someone to take real liability when things go wrong. “AI did it” doesn’t fly with customers or regulators.
Production edge cases, integration messes, architecture trade-offs, and long-term ownership still demand real human judgment.
@GaryMarcus The only guardrails that work are hard rules where you physically do not let the AI execute certain commands.
Everything else is likely to be ignored eventually.
@IroncladDev I’ve always hated PRa from both the reviewer and submitter perspective. They are super boring and require context switching into whatever the PR is about.
I understand all the purposes of them, but they still suck. Unfortunately I do not have a better alternative