How to keep thinking

This title could be clearer and more informative.Try out Clickbait Shieldfor free (5 uses left this month).

Working with AI agents in 2026 has shifted software engineering toward rapid context-switching and skimming LLM outputs, leaving less time for slow, deliberate thinking. The author argues this risks atrophying deep reasoning skills and recommends two countermeasures: writing in your own words (not with LLMs) to force genuine articulation of ideas, and reading dense non-fiction books slowly. Combining both — reading a book and then writing about it — is presented as the most effective way to keep the habit of slow thinking alive outside of work, which remains necessary for tasks like large refactors that current LLMs can't yet handle tastefully.

6m read timeFrom seangoedecke.com
Post cover image
Table of contents
Why not slow down?To keep on thinking, read and writeDon’t lose the habit

Questions this post answers

Does using LLMs for coding make you worse at deep thinking over time?

Heavy LLM use biases developers toward fast 'skimming and judging' mental patterns and away from the slow 'hammock time' needed for deep thought and creativity. The concern is not literal cognitive damage but a gradual shift in mental habits. Tasks like large refactors on complex codebases still require sustained independent reasoning, and LLMs cannot yet perform them tastefully without human judgment. Developers navigating this trade-off share their experiences on daily.dev, where the AI-assisted development conversation keeps moving.

What can I do to keep my deep thinking skills sharp while using AI coding tools heavily?

Writing in your own words — not with an LLM — forces you to articulate and construct ideas rather than just evaluate them. Reading dense non-fiction books slowly provides information-dense content that counteracts the shallow skimming that LLM workflows encourage. Combining both by reading a book and then writing about it is especially effective for exercising the slow, deliberate reasoning that AI-assisted work tends to crowd out. Engineers working through the same tension between speed and depth discuss it on daily.dev.

327.1K Impressions11 Comments