為轉售商開發了AI驅動的利潤計算器和庫存追蹤應用程式

So I built Underpriced.app - an AI reseller app that tells you if something is worth flipping before you buy it.

The Problem

Researching items takes ~5 minutes per item
You waste time and miss deals while researching (someone else grabs it)
You buy blindly and hope for the best
Spreadsheet tracking is tedious

The Solution

Take a photo or screenshot → AI identifies the item → Get instant analysis:
Profit potential after all fees
ROI percentage
Demand level & time to sell
Best platforms to sell on
Red flags to watch for
Deal score (0-100)

Tech Stack

Frontend: Next.js, React, TailwindCSS
Backend: Node.js, PostgreSQL
Chrome Extension for browser integration

Features I'm Proud Of

Works for both online sourcing (screenshots) and in-person (photos)
Built-in flip tracker with analytics & reports
Chrome extension for one-click online deals analysis
Premium AI tools: flipping strategy, listing generators, deep market research
15-30 second analysis time

Current Status

Launched 2 months ago
10 free analyses to start
Paid tiers from $2.99/mo
Active users in flipping community

What I'm Working On
More accurate valuations for niche items
Multi-item batch analysis
Mobile app (currently PWA)
Integration with inventory management tools

Lessons Learned
AI vision models are incredible but need heavy prompt engineering
Promoting SaaS is way harder than building it
The reselling community is way bigger than I expected
Balance between speed and accuracy is crucial

Try it: underpriced.app

Would love feedback from fellow builders:

How would you improve the AI prompts for better accuracy?
What's a fair price for this kind of analysis tool?
Should I focus on depth (better analysis) or breadth (more features)?
Happy to answer any technical questions about the build!