A wholesale supplier selling into e-commerce · Wholesale and e-commerce

Only the profitable rows come back.

Drop a supplier spreadsheet in, get back only the rows worth buying, with the lookup and the margin math already done.

What is happening now.

Luxin, an AI consulting firm, built, is building, or proposed this work. The status above says which.

Two to three minutes to look up each product code by hand, roughly 25 a day.
Case-pack codes, trimmed digits and missing listings, each one a judgement call.
Supplier lists of 10,000 rows skipped entirely because they are too long to work through.

2–3 minutes: per manual product-code lookup, as the problem found; The problem found
10,000 rows: a supplier list too long to work through by hand; The problem found
$500: the fixed price of the pilot, one payment; What is proposed

Every figure says where it came from. A problem found is not a result.

01
A supplier spreadsheet dropped into a small Claude project on the owner’s own account.
02
Each product code looked up in the pricing tool by the browser extension, with the case-pack math and the digit fallbacks handled.
03
Every row scored against the owner’s own rules. Only the profitable ones come back.
04
The owner approves or rejects each find, and the system learns his taste from the answers.

The tools it works inside.

Illustration of the pattern. Not a screenshot, and not real client data.

The list is prepared. A person checks every row before anything is bought or sent to a client.

Proposed

Proposed, not yet delivered. Nothing on this page is a result.

Thirty minutes, no deck. You describe how the work actually happens, and I tell you plainly whether there is something worth building.