Predicting Inventory

Working with Google's retail industry team to build an digital demonstration for inventory management for retailers.

Users access a mobile web app through a device, and are given a recipe to shop for.

With real products pulled from an Instacart dataset, as users browse and add items to cart, custom recommendations are found using BQML based on 20gb of historical shopping data that was generated.

As users shop using the web app, Cloud Spanner and BigQuery databases are updated and inventory is pushed to Firebase for realtime access. A warehouse manager dashboard reflects these changes, and provides a mechanism for tracking inventory and receiving suggestions for couponed items.