I built an AI Agent in 2 hours (and got paid $2600)

I built an AI Agent in 2 hours (and got paid $2600)

by Nate Herk

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Nate Herk 👥 964K 📅 December 17, 2025 ⏱ 27 min 👁 40K 📄 tutorial 🧭 2026-08-28
Available in: English (current) Français

Keywords

AI agentn8nautomationclientworkflow

Summary

In this video, Nate Herk demonstrates how he built an AI agent for a client in just two hours and was paid $2,600. He starts by explaining the agent’s purpose: to automate the onboarding process for new students of a course. The system consists of four main workflows: creating a CRM entry when a payment is made, sending a follow-up email after three days if the student hasn’t created an account, escalating to a human after five days, and finally sending a personalized welcome email using AI once the account is created. Nate walks through the entire build in n8n, showing how to set up webhooks, Gmail, Slack, Google Sheets, and an AI agent with a structured output parser. He emphasizes that the first three workflows use no AI, making them robust and reliable, while the final one uses AI for personalization. He also discusses why the client paid $2,600, attributing it to the value of saving time and improving customer experience. The video concludes with advice on how to replicate this success, including tips on selling AI agents and the importance of understanding the client’s needs.

187 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides high practical value for viewers interested in building AI agents for clients. It offers a real, detailed example of a client project, including the wireframe, the build process, and the reasoning behind each step. The argumentation is solid: Nate explains why the automation is valuable (saving time, reducing manual follow-ups, improving onboarding experience) and justifies the price by the efficiency and reliability of the system. He also addresses potential concerns, such as the importance of testing and the limited use of AI to ensure robustness. The tutorial is actionable, with clear explanations of n8n nodes and configurations.

Scientific Rigor, Source Quality, Title Accuracy

The video is a tutorial and case study, not a scientific presentation. It does not cite external sources, but it does reference the tools used (n8n, Gmail, Slack, Google Sheets, Claude Sonnet 4.5) and provides links to his courses and services in the description. The title accurately reflects the content, and the video delivers on its promise. The creator is transparent about the build process and the client payment, which adds credibility, though the promotional nature of the video (links to paid courses) should be noted.

201 words

Title / Content Match

The title accurately reflects the content: the creator builds an AI agent in 2 hours and discusses the $2600 payment, with a detailed walkthrough of the build.

Quality & Reliability

7/10

The video provides a transparent, step-by-step tutorial on building an AI agent using n8n, with clear explanations of the workflow logic and practical implementation. The creator demonstrates real client work and shares the actual build process, which adds credibility. However, the video is primarily promotional, with links to paid courses and tools, and lacks external scientific sources or rigorous validation of the claims.

Chapters

Cited Sources

Concurring Sources

  • n8n documentation — The video uses n8n for automation; the documentation confirms the features and usage.
  • Claude API documentation — The video uses Claude Sonnet 4.5; the documentation provides details on the model.

Contribution & Novelties

The video offers a practical, real-world example of building an AI agent for a client, showing the entire process from wireframing to implementation. It emphasizes the importance of using AI only where necessary, which is a nuanced approach that adds value. The tutorial is accessible for beginners and provides a clear template for similar projects.

Pour aller plus loin :

101 words

Radar Profile

The radar profile shows high scores in quantity of information and technical level, reflecting the detailed tutorial. Quality of information is also strong, but reliability is slightly lower due to the promotional nature and lack of external validation. The overall balance indicates a practical, hands-on resource rather than a rigorous scientific analysis.

Reliability 6/10

💬 Très positif. Sur les 30 commentaires analysés, la majorité exprime de l'enthousiasme et de l'intérêt pour le contenu, avec des questions techniques et des remerciements. Quelques commentaires critiques soulèvent des points sur la viabilité du modèle d'agence, mais restent constructifs.