Case Study: Labelbox boosts AI development with Elastic on Google Cloud

Labelbox enables customers to build better AI products with Elastic Search on Google Cloud

Labelbox, an AI platform that helps customers prepare labeled data for machine learning, needed a faster and more flexible way to search and analyze unstructured data in its Catalog product. Its existing PostgreSQL-based setup made it hard to build accurate filters and could take up to 20 seconds to return search results, slowing AI development for customers like Procter & Gamble, Genentech, and Walmart.

Elastic implemented Elasticsearch on Elastic Cloud on Google Cloud to power Labelbox Catalog with scalable search, richer analytics, and easier deployment. The result was a major improvement: search response times dropped from 20 seconds to about 1 second, customers began uploading more data and requesting more filters, and Labelbox saw increased usage and growth with some of its biggest customers.

Labelbox

Matthew Chambers

Principal Engineer