I Cloned Calendly and Now It’s Free Forever

I Cloned Calendly and Now It’s Free Forever

🎙 Nate Herk 👥 965K 📅 August 28, 2026 ⏱ 14 min 👁 469 📄 tutorial 🧭 2026-08-28
Available in: English (current) Français

Keywords

AI agentSaaSCalendlyOpenAI Codexvibe coding

Summary

Nate Herk presents SnagTime, a scheduling application he built using AI agents (primarily OpenAI Codex) over five days, claiming it replicates core Calendly features. The video includes a live demo of the app, showing event types, calendar syncing, Stripe payment integration, and team permissions. He details the build process, which involved five main prompts: a goal prompt, a rebrand/redesign prompt, and performance enhancement prompts. The project consumed 334 hours of agent runtime, 76 unique sub-agents, and 32 billion input tokens, costing an estimated $15,000 in inference, but was covered by his $200/month Codex plan. He emphasizes that while the initial build was efficient, scaling and maintaining the product requires ongoing human oversight and prompt-based fixes. The video also promotes his free Skool community and other resources, and includes a sponsorship segment.

131 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video’s primary value lies in its concrete, first-hand data on the cost and effort of AI-assisted software development. The creator provides specific metrics (token counts, agent hours, dollar costs) that are rarely shared, offering a realistic benchmark for similar projects. The argumentation is straightforward and practical, focusing on the trade-offs between using AI agents and traditional development. However, the claims are based on a single, uncontrolled case study, and the creator’s dual role as a promoter of AI automation services introduces a potential conflict of interest. The demonstration is convincing but not independently verified.

Scientific Rigor, Source Quality, Title Accuracy

The video does not cite external scientific or technical sources; it relies on the creator’s own experience and the tools he used. The description includes links to his Skool community, a voice-to-text tool, and a VPS hosting service, which are promotional rather than authoritative references. The title is accurate but slightly misleading, as the ‘free forever’ claim is qualified by the need for ongoing maintenance and potential costs. The video’s internal consistency is good, but the lack of external validation limits its scientific rigor.

194 words

Title / Content Match

The title accurately reflects the content: the creator demonstrates a Calendly clone and claims it is free to run, though the video clarifies that maintenance requires ongoing effort.

Quality & Reliability

6/10

The video provides a detailed, first-hand account of building a scheduling app with AI agents, including specific metrics (time, cost, token usage) and a live demonstration. However, it lacks independent verification, relies on anecdotal evidence, and contains promotional elements for the creator's community and tools.

Chapters

Cited Sources

Concurring Sources

Contribution & Novelties

The video contributes a rare, quantified case study of building a functional SaaS application with AI agents, providing concrete data on token usage, agent hours, and cost. It highlights the gap between AI-generated code and production-ready software, emphasizing the need for human UX fixes and ongoing maintenance. The ‘five-prompt’ methodology offers a practical framework for similar projects.

Pour aller plus loin :

  • OpenAI Codex — The AI tool used for the build; understanding its capabilities is central to the video’s claims.
  • Cal.com — An open-source Calendly alternative, relevant for comparing features and costs.
  • Vibe Coding — A term describing the AI-assisted development approach used in the video.

107 words

Radar Profile

The radar profile shows a balanced but moderate performance across all dimensions. The video scores highest on information quantity and technical level, reflecting the detailed metrics and technical walkthrough. However, information quality and reliability are lower, due to the lack of external sources and the promotional context. This suggests a practical, hands-on tutorial with limited scientific depth.

Reliability 5/10