Case studies
A pre-owned luxury watch marketplace
An AI sales agent built inside the owner's existing sales channels.
E-commerce, client anonymised
Context
A pre-owned luxury watch marketplace that sells through its website and messaging channels.
How we worked
Embedded with the owner. We built inside the existing sales channels instead of adding a new one, and measured how the agent handled conversations.
What we built
- A website rebuild on Next.js, Sanity, FastAPI and Supabase
- A WhatsApp and Telegram AI sales agent on Claude, with a channel-adapter pattern so one agent serves both channels
- Inventory scraping and sync
- Webhook reliability, CDN and bot mitigation
The loop
- Find. Worked with the owner on how sales conversations actually run.
- Build. Chose a channel-adapter design so one agent serves WhatsApp and Telegram.
- Ship. Put the agent live in the existing sales channels.
- Adopt. Measured how the agent handled conversations.
Outcome
Outcome figures are being confirmed with the client before we publish them.
Stack
Next.js, Sanity, FastAPI, Supabase, Claude, WhatsApp, Telegram
Let's find where AI fits in your business.
Book a discovery call. We'll tell you where AI would pay off in your systems, and what the first build would take.