Reading 40,000 shops is only half the job. The other half is making sense of it: spotting the same sneaker in dozens of shops, remembering what it cost last month and letting an AI agent find all of it in one query. You design that data layer together with the founder. It runs on Postgres and holds 21.5 million product variants today.
Design the data model for 40,000 shops: products, sizes, prices, stock and history Build the checks that keep the data fresh, and the matching that recognises one product across shops Shape the index that AI agents can query, and measure whether they use it
You think in tables, keys and queries, and your SQL is solid A wrong number bothers you, and you can prove which one is right You are curious how AI models read structured data: feeds, APIs and schemas
#J-18808-Ljbffr
