Case study
AI that earns money in a real resale business
A multi-channel resale platform where AI writes the listings, cleans the photos and drives the marketplace, in production for a trading business.
Channelytix (my own product)Retail and e-commerce
Built from conceptDelivered
-, delivered in In production
The problem
Listing second-hand stock is slow, repetitive work: photos, titles, descriptions, prices and paperwork for every item.
How I approached it
- AI features built as plugins, so models can change without rewrites
- An MCP server so AI assistants can drive orders, shipments and invoices safely
- Quality gates on every change: code scanning, security analysis and reviews
What was built
- AI-written listing titles and descriptions
- Automatic photo background removal
- Marketplace listing automation
- Purchase orders, sales orders, shipments and invoices
Stack
- .NET 10
- Angular
- Azure Container Apps
- Azure AI Foundry
- MCP
Outcomes
- Listing time cut from minutes to seconds per item
- Sale-to-invoice fully automated
Next step
Want the same result on your problem?
In 30 minutes you'll know whether to fix it, buy it, build it or leave it. No pitch deck.