GoKawiilAlex Ewerlöf argues AI coding tools haven't solved software engineering
Developer and author Alex Ewerlöf published an essay disputing the claim that 'coding is solved' by AI. He argues that while LLMs make code creation cheaper, they haven't addressed non-functional requirements like maintenance, reliability, security and scalability, which he says make up most of the real cost of production software. He also contends that even basic functional correctness remains unsolved by current AI coding tools.
GoKawiil's interpretation of the reporting above, not reported fact.
The piece pushes back against a popular narrative in tech circles that AI has reduced engineering to matters of 'taste,' arguing instead that most commercial software still demands rigorous verification that current tools don't provide. Ewerlöf's framing suggests that overconfidence in AI-generated code may stem from users not carefully reading its output, a claim he presents as his own observation rather than a proven finding.
- Ewerlöf distinguishes between low-risk use cases (personal tools, prototypes) where unreviewed AI code may be acceptable and high-risk production systems where it is not.
- He argues non-functional requirements like security and scalability, not just code generation, dominate real engineering costs.
- The essay challenges the widely circulated claim that coding as a discipline has been 'solved' by AI tools.
Source: blog.alexewerlof.com — Alex Ewerlöf, 2026-09-28
Published there as: “Coding Is Not Solved – Alex Ewerlöf Notes”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.