Work Drapeur

Custom AI engineering in practice

Drapeur: from ChatGPT to listings ready for review.

Drapeur’s problem went beyond generating text. Product listings needed to be prepared inside the merchant’s tools, with review before publishing. This project shows how I work: understand the process, build the agent, and connect it to the existing tools.

See the role, results, and boundaries

The business problem

Product listings ready for review.

This merchant already used ChatGPT. They still had to copy the content and create every listing in Shopify. The agent took over that preparation inside the store.

For 20 to 30 daily listings, the client estimates saving one to two weeks of work per month. Listings remain in draft until the merchant approves them.

I built and installed the agent for this workflow. The client uses it independently and continues to refine the instructions.

Prepare, then review
Copy + visuals
ShopifyDraft

You stay in control of publishing.

“It creates all my product listings for me. […] So a huge thank you, it saves me an incredible amount of time.”

AlexisCEO, DrapeurTranslated from French

How this project works

The agent prepares. The merchant decides.

01

The merchant’s guidelines

The client’s criteria guide how copy and visuals are prepared.

02

Preparation by the agent

The agent fills out product listings in Shopify without publishing them automatically.

03

The merchant’s approval

The client reviews the drafts and decides what to publish.

This case describes the workflow delivered for Drapeur. Another project needs its own scope.

My approach to AI consulting and engineering

Your next step

What problem would you like to solve?

Tell me about your business, what gets in the way, and the result you need. The first 30-minute conversation is free. We’ll see whether I can help and where to start.

Let’s talkBook a free first conversation

Scope, timeline, price, and the expected outcome are defined in writing.