Les devs ne codent presque plus en 2026

Les devs ne codent presque plus en 2026

Developers will hardly be coding anymore by 2026

Title history
  • Les devs ne codent presque plus en 2026 — observed on 20261101-0148 UTC (YouTube)
🎙 DevArt 👥 19K 📅 January 20, 2026 ⏱ 14 min 👁 8K 📄 expert opinion 🧭 2026-11-01
Available in: English (current) Français

Keywords

code reviewAI-assisted developmentpull requestvibe codingdeveloper evolution

Summary

The video addresses the shift in the developer role in 2026, emphasizing that developers now spend more time reading and reviewing code than writing it from scratch. The creator shares insights from personal experience and colleagues, confirming that the volume of pull requests has multiplied due to AI tools like Claude Code. He references the State of Code report by Sonar, which highlights that the most important skill is now reviewing and validating AI-generated code for quality and security, followed by efficient prompting. The video explains that while the fundamental knowledge of software engineering remains necessary, the ability to create components from memory is less critical. He advises juniors to train by comparing different LLM outputs on the same task to develop review skills. He also discusses tool costs and the importance of making pull requests manageable for effective review. The tone is reflective and reassuring, asserting that the profession is evolving rather than disappearing, and concludes with a call to action to like, subscribe, and prioritize family.

168 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video offers a valuable first-hand perspective on the daily reality of developers in 2026, supported by a referenced industry report. The argumentation is convincing in describing the increased emphasis on code review and the shift in required skills. However, it relies heavily on personal anecdotes and does not provide quantitative evidence beyond the cited report. The reasoning is coherent and practical, suggesting concrete training methods for juniors, but it remains a subjective viewpoint rather than a comprehensive analysis.

Scientific Rigor, Source Quality, Title Accuracy

The video is grounded in personal experience and a single referenced report (State of Code by Sonar), which adds credibility. However, the source is not deeply analyzed, and no contrasting data or studies are presented. The title accurately matches the content, which is a professional reflection on the changing role of developers. The creator does not cite multiple independent sources, limiting the scientific rigor. The comments tend to confirm the phenomenon but also express concern, aligning with the video’s narrative.

174 words

Title / Content Match

The title accurately reflects the central claim that developers spend less time coding and more time reviewing AI-generated code.

Quality & Reliability

6/10

The video is based on personal experience and a referenced industry report (State of Code by Sonar), but lacks detailed methodological context or verification of claims. It offers a credible qualitative perspective on the changing developer role, but should be considered anecdotal rather than rigorously evidenced.

Key Moments

Cited Sources

Concurring Sources

  • State Of Code by Sonar — The report supports the claim that reviewing AI-generated code is the number one skill.

External References

Contribution & Novelties

The video provides an original first-person account of the practical impact of AI on the daily work of a developer, emphasizing the shift from coding to code review. It offers a concrete training suggestion for juniors: compare outputs from different LLMs on the same task to develop evaluation skills. The reinforcement of the importance of review skills over raw coding is a valuable insight for the community.

Pour aller plus loin :

109 words

Radar Profile

The profile shows a moderate balance across all dimensions, with slightly higher scores in information quantity and technical level, but lower in reliability due to the anecdotal nature and single-source dependence. This suggests a persuasive but not deeply evidence-based presentation.

Reliability 5/10

💬 Équilibré avec une tonalité nostalgique. Sur les 30 commentaires analysés, le climat est partagé entre regret du codage manuel et acceptation de l'évolution, avec une majorité de retours reflétant une appréhension quant à l'avenir du métier.