Salesforce Einstein: Machine Learning in a Multi-Tenant Environment
A typical Machine Learning pipeline involves a series of steps to build a single use model from a predefined set of product features. Yet we at Salesforce need to build models for hundreds of thousands of customers who have very different behaviors and data. Join us as we share the challenges we faced building Machine Learning in Salesforce Einstein. Learn how we've automated the most repetitive Machine Learning tasks, like feature generation, model selection, and hyper-parameter tuning and optimization, to build hundreds of thousands of models, while delivering quicker modeling turnarounds and higher accuracy.