J'ai arrêté de coder : Cette IA fait TOUT à ma place (démo)

J'ai arrêté de coder : Cette IA fait TOUT à ma place (démo)

I've stopped coding: This AI does EVERYTHING for me (demo)

Title history
  • J'ai arrêté de coder : Cette IA fait TOUT à ma place (démo) — observed on 20261112-0057 UTC (YouTube)
🎙 Thomas Berton 👥 48K 📅 December 4, 2025 ⏱ 23 min 👁 7K 📄 tutorial 🧭 2026-11-12
Available in: English (current) Français

Keywords

Manus AIAI agentwebsite builderFigmaCRM

Summary

In this 23-minute video, Thomas Berton demonstrates how Manus AI v1.5, an autonomous AI agent, can create a professional portfolio website from a Figma template without writing code. He walks through the process step-by-step, from selecting a design on Figma and crafting an optimized prompt, to launching the agent, customizing the design, deploying the site with a custom domain, and adding a lead capture form with a backend CRM and click analytics. The video highlights features such as real-time editing, voice input prompts, deep links for social media, and direct integration with external services via MCPs. Thomas emphasizes the speed and ease of the process, claiming the entire project is built in ‘one shot’ and deployed in minutes. He also promotes his own community and offers a free trial for Manus. The video is essentially a product demonstration rather than an independent review, with the creator clearly affiliated with the tool.

151 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides a clear, practical demonstration of an AI agent’s capabilities for web development, which is valuable for viewers interested in no-code solutions. The argumentation is based on a live demo rather than abstract claims, enhancing credibility. However, the content is largely promotional: the creator repeatedly emphasizes advantages without discussing limitations, potential pitfalls, or costs, and encourages joining his paid community. The reasoning is anecdotal and lacks comparative analysis with other tools. The claim of ‘one-shot’ success is presented without addressing edge cases or variability in AI performance, which could mislead less experienced viewers.

Scientific Rigor, Source Quality, Title Accuracy

The video cites no external scientific sources and relies solely on the creator’s own experience and promotional materials. The description includes affiliate links and a free-trial redemption code, confirming a financial stake in convincing viewers to adopt Manus. The title accurately reflects the content, and the video quality is professional. As the creator is not an AI researcher but a practitioner, the technical depth is limited to user-level instructions. There is no analysis of underlying AI mechanisms, security, or data privacy. The adéquation titre is good, but the overall rigor is weak due to lack of objectivity and absent external references.

211 words

Title / Content Match

Title accurately reflects content: the creator demonstrates stopping manual coding by using Manus AI to build a complete website.

Quality & Reliability

6/10

The video is a product demo emphasizing practical results but lacks critical analysis, independent verification, or comparative benchmarks. The author is a promoter (affiliate links, premium community) which introduces bias. However, the demonstration is detailed and reproducible, and the claimed features are consistent with current AI agent capabilities.

Chapters

Cited Sources

Concurring Sources

Contribution & Novelties

The video showcases Manus AI’s integrated capabilities (databases, deployment, CRM functions) in a single workflow, which is a novel demonstration of autonomous agents for non-coders. It offers practical insights into prompt optimization and using Figma templates as design seeds. However, it adds little new technical knowledge.

Pour aller plus loin :

97 words

Radar Profile

The radar profile shows moderate scores across all dimensions, with quantite_information slightly higher than qualite and fiabilite. This indicates that while the video provides a substantial amount of practical information, its reliability is diminished by promotional bias and lack of independent verification.

Reliability 5/10