Totally fair. You can build the first 10-minute demo with an LLM now.
The hard part is usually what comes after: editable state, templates, email-safe HTML, PDF/export pipelines, assets, permissions, custom blocks, previews, and all the rendering edge cases that show up once real users depend on it.
AI also makes structure more important, not less. That’s why we launched Unlayer Elements: instead of having an LLM generate a static blob of HTML, it can target structured components that map back into an editable email, web page or document.
So I'd use AI to prototype it. I'd use Unlayer when that content layer becomes production infrastructure you don't want to maintain yourself.
As a content creator, this intrigued me until I saw the AI-generated overview video front and center on the homepage. The subtext communicates that the product was designed for fake people standing in front of large fake touchscreens dragging fake elements around.
Fair critique on the video :') We’re working on a studio-recorded version, but didn’t want to delay the launch, so we hacked this together as the script/placeholder for now. The demo is completely, 100% recorded in blood and flesh of a human: https://www.youtube.com/watch?v=0HsDtNkdMpM
The hard part is usually what comes after: editable state, templates, email-safe HTML, PDF/export pipelines, assets, permissions, custom blocks, previews, and all the rendering edge cases that show up once real users depend on it.
AI also makes structure more important, not less. That’s why we launched Unlayer Elements: instead of having an LLM generate a static blob of HTML, it can target structured components that map back into an editable email, web page or document.
So I'd use AI to prototype it. I'd use Unlayer when that content layer becomes production infrastructure you don't want to maintain yourself.
I work outside of SV in the southeast and most businesses know AI is being used for products they use but they want experts in that field behind that.
Sorry, but that's a no-go at this point.