The rise of judgement over technical skill

Management, leadership, and judgment

• Some argue great leaders start as practitioners whose hands-on skills fade but whose judgment scales impact.
• Others contest the idea that good engineers are routinely promoted to management; when it happens, it often reflects combined technical and people skills plus explicit management training.
• Several comments stress that management skill is orthogonal to technical skill, similar to how subject expertise differs from teaching ability.

Judgment vs. technical skill

• Core tension: can you have good judgment without deep technical skill? Many say no—judgment about code, systems, or art rests on years of doing.
• Judgment is framed as knowing what to build, why, and when something is off; skill is needed to diagnose and fix what’s wrong.
• Others argue AI could eventually surpass humans in some aspects of judgment by incorporating formal statistical reasoning, though current systems struggle to revise assumptions.

AI coding assistants in practice

• Multiple commenters report that the bottleneck has shifted to:
  • Problem decomposition into small, well-bounded tasks.
  • Exhaustive code review and integration.
• Some find AI dramatically accelerates exploration, refactoring, and “gold-plating” infrastructure; others find it wastes time, produces incorrect or obsolete code, and never reaches “senior engineer” competence.
• A common analogy is “unlimited junior interns who never really improve”: useful for well-specified subtasks, but high review overhead and no compounding returns.
• Flow and enjoyment: several experienced engineers say AI tools interrupt concentration, and they prefer writing code themselves.

Impact on juniors, learning, and education

• Concern that juniors may over-trust AI output and stunt their own skill development; some companies prohibit or heavily constrain LLM-generated code for junior staff.
• Parallels are drawn to education debates: critical analysis/judgment is meaningless without broad foundational knowledge and mental models.

Offshoring, labor, and productivity claims

• Historical attempts to “cheap out” by offshoring are cited: you can hire many low-cost devs, but coordination, quality, and judgment remain the real constraints.
• Some see AI as the next iteration of this—another way to push routine work down—but emphasize that experienced “drivers” become more valuable, not less.
• Others note that current layoffs are more clearly tied to offshoring and macroeconomics than to AI.

Art, music, and creativity analogies

• Music and design: AI can reach “superficially professional” output, but commenters argue that real quality and originality still require taste and practice.
• Some say democratization has shifted which skills matter (e.g., DAWs instead of instruments), not replaced skill with pure judgment.

Meta-critique of the article and AI hype

• Several readers find the piece thin, mostly restating an Eno quote and riding a “AI changes everything” narrative.
• There’s frustration with overuse of terms like “democratization” and with essays that loosely assert “tooling is solved; only judgment matters” without showing concrete evidence.