Using De-duplication, Fuzzy Matching & Schema Unification
How do you help retailers, reduce supplier duplicates by 67%, enable better spend visibility and performance scoring and improve onboard automation. Review the architecture we used and the script to train a Supplier de-dup model
Using Embeddings to Clean and Group Product Catalogs
How many times has a retailer struggled with inconsistent product listings. For example, the product listing says Blue Women’s Polo Shirt - L vs, Women Polo Tee Large Blue. There are also taxonomy differences like Apparel instead of Tops/Tees and sometimes missing metadata. These inconsistencies can hurt Search, personalization and optimization. Lets show you how you can tackle this with a transformer based embedding
Prompt Engineering for Retail Agents: Lessons from the Field
Real insights from using LLMs in Customer Support and Procurement
Designing a Clean Data Lake for Retail Supply Chains
Retailers that are managing multiple brands face messy data due to disparity in naming conventions, format challenges, duplicate or redundant product records and unreliable timestamp and stock information.
This blog walks through the framework we built to create value through a clean Data Lake for Retail Supply Chains