Comment by kevinreisenbuk
3 months ago
I’m building an AI-powered fashion search engine that helps people find clothes that actually match their style, fit and price across 1000s of brands and millions of products
Most shoppers spend hours to find the rights product. We’re fixing that with intent-based search that understands descriptions, images and personal preferences.
We’ve hit 25K+ searches in 4 months, growing 50% MoM, and built our own scraping system that makes product data collection 100× cheaper than existing tools.
Still early, but live. Would love feedback on search quality and result relevance.
PS! There are some products out of stock, this is expected, fixing it right now.
Really cool. Found myself a specific jacket that I have been looking for a long time and made the purchase. 25k+ searches in 4 months. Which markets you launched in or marketed so far? How you chose where to start?
pretty damn cool. tested it with some still frames from movies and pinterest boards and it found most of the things.
[stub-for-green-comments]
[dead]
Thanks! We’re adding millions of products and thousands of new brands over the next few weeks. We’re also exploring second-hand listings.
[dead]
We differentiate by focusing on intent, not keywords. The giants return results based on matching terms. We return results based on understanding what the shopper actually wants to do, backed by curated, high-quality data.
[dead]
Biggest challenges in search relevance are: - collecting and maintaining a large, clean, diverse product catalog (up to hundreds of millions of products) - actually understanding what users mean (not just what they type) - giving them ways to refine vague searches.
Skip any of these and even the fanciest ranking algorithms feel useless. Helping users bridge that gap is where relevance actually clicks.
[dead]
[dead]
[dead]
[dead]
[flagged]