Launched an AI product photo tool after being frustrated with existing solutions

Hey IH! 👋

I just launched ShotDirector — an AI tool that generates professional product photos in 30 seconds.

The Problem I Was Solving:

I was spending 2-3 hours on AI tools like Midjourney trying to get decent product photos. Complex prompts, endless regenerations, mediocre outputs. It was supposed to be "instant" but felt like pulling teeth.

I thought: "Why do I need to be a prompt engineer just to get a good product photo?"

What I Built:

• Upload one product photo
• Get 5 cinema-quality shots instantly
• No prompts, no technical skills needed
• Consistent, brand-aligned results every time

Tech Stack:

• Frontend: Next.js 14 + TypeScript
• AI: Custom-trained models on top of Stable Diffusion
• Hosting: Vercel
• Database: Supabase

Current Status:

• Built in 3 months
• Soft launched 2 weeks ago
• ~50 users testing
• $0 in revenue (free beta for now)

What I Learned:

1. Solving your own problem is powerful. I use this daily, which means I know exactly what's broken.

2. Speed matters more than perfection. I shipped an MVP with 60% of planned features. Users don't care — they just want it to work.

3. AI tools are still too complex. Most people don't want to learn prompt engineering. They want results.

4. The "better AI tool" positioning works. Instead of competing with traditional photography, I'm competing with other AI tools by being simpler and better.

Next Steps:

• Add batch processing for multiple products
• Launch paid tier ($29/mo)
• Build out style customization

Questions for IH:

• How did you price your first AI tool?
• Should I focus on B2C (founders) or B2B (agencies)?
• Any tips on getting initial paying customers?

Would love feedback! 🙏