I keep running into the same friction with creators I know. They have a clear vision for a music video or social short, but translating that into production-ready prompts for Kling, Runway, or Pika is where things fall apart. The creative brain and the prompt-engineering brain seem to work differently, and most solo creators aren't in the business of becoming prompt engineers.
That's the gap I'd look at: a conversational interface that sits between the creator's vision and the generation models. Describe what you want naturally—"late-night Tokyo street scene with neon reflections, melancholic vibe"—and the tool handles the structured prompt output, scene breakdowns, and iterative refinement through dialogue. The product lives entirely in the UX layer, not the generation layer itself.
The target user is pretty specific: solo music video directors, indie artists, and small social media agencies who produce video content regularly. The emotional hook is simple—you feel like a director, not a prompt monkey.
What makes me pause is whether this is actually a hard problem or just a UX convenience that gets commoditized quickly. And the dependency on underlying model quality is real. But the shareable output creates a built-in growth loop—every AI-directed music video posted online is a soft ad. Plus, at $29–$79/month, you're competing with $500–$2,000 per-project creative director rates, which anchors willingness to pay pretty well.
Am I overthinking the prompt problem, or is this a real bottleneck worth solving?
Top comments (0)