The Programmer Identity Crisis - Simon Hojberg - Principal Frontend Engineer

Programmers are losing their identity as creators and becoming operators for AI agents. The article argues that large language models and "vibe-coding" replace deep engagement with shallow, prompt-driven output, stripping away the craft that defines the profession. It shows how the promise of faster productivity masks a loss of agency and critical thinking.

Management is increasingly mandating LLM usage, forcing developers to adopt standardized prompts and specifications instead of personal toolchains. The author describes how custom editor configurations, dotfiles, and immersive coding sessions are being replaced by corporate-chosen AI workflows, turning a once-proud, self-curated practice into a dictated process.

The consequences are concrete: cognitive debt grows as developers skim AI-generated code without understanding, reviewers become the first line of defense against hallucinated libraries and broken logic, and morale drops when the joy of solving puzzles is replaced by repetitive operator tasks. This erosion of craft leads to lower code quality, higher bug rates, and a workforce that feels disconnected from its work.

Technical leaders must recognize the trade-off and protect the space for deep, immersive coding. AI should handle repetitive chores, not replace the mental model building that underpins reliable software. By setting clear boundaries, encouraging personal tool mastery, and preserving time for focused craftsmanship, leaders can keep teams productive, engaged, and resilient in the face of AI hype.