Why it matters
This demonstration marks a shift in software development where the 'infrastructure' of a SaaS is no longer a high-bar bar...
I Cloned Calendly and Now It’s Free Forever
Nate Herk | AI Automation
The Signal
A developer recently demonstrated a functional, Calendarly-style scheduling application named "Snag Time" built entirely through agentic AI coding over five days. The project showcases how multi-agent iterative loops can produce complex software, yet it also highlights significant gaps between a functional local prototype and a production-ready, feature-complete SaaS product.
The Case
The Build Process
- The application resulted from a massive multi-phase prompt—research, planning, building, and testing—executed by 76 unique sub-agents over 334 total agent-hours.
- The author reports the project consumed over 32 billion input tokens and 47 million output tokens, resulting in an estimated $15,000 in total inference value, though it cost only $150 out-of-pocket by leveraging a $200/month subscription.
- The build required extensive iterative test-and-fix loops, which the author cites as the primary engine for functionality rather than a one-shot successful prompt.
Product Capability and Limits
- The demo successfully executes live calendar-state synchronization, including blocking availability after a manual "busy" entry, and processes payments via a sandbox Stripe flow.
- Despite these features, the author concedes the app is currently a localhost-only prototype that lacks the deep automation functionality standard in commercial tools like Calendarly.
- Agentic testing failed to identify intuitive UX issues, such as non-clickable progress bars that humans expected to interact with, forcing the author to perform manual interface redesigns.
The 1 Minute Signal Take
While agentic AI can generate sophisticated internal tools at a fraction of the traditional cost, the author's experience demonstrates that such builds require manual oversight, performance tuning, and ongoing maintenance. This is a functional prototype for internal use, not a "set-and-forget" replacement for commercial SaaS infrastructure.
Pro Analysis
Full analysis always available on Pro.
Time saved:
Tags
Written by: 1 Minute Signal Editorial Team