In the previous chapter, Salesforce Einstein Vision, we covered Einstein Vision APIs and how one can build web-based applications on Heroku. We used Node.js and React as our technology stack. This chapter focuses on how one can leverage the Salesforce Force.com platform to integrate with Einstein Vision APIs and provide image recognition capability across the sales, service, and Community Cloud. We will be leveraging Apex for all our business logic and outbound API callouts and Lightning Components to build user interfaces. In the previous chapter, Salesforce Einstein Vision, we assumed that the training of dataset will be carried out by the administrator via the command line. A shell script can be written on Linux, but we can also leverage the Force.com power to automate the training of dataset natively on the platform.

In this chapter, we will cover the following topics:

• Setting up authorization between Salesforce and Einstein Vision APIs
• Creating and training a dataset via Apex
• Creating an administration app to create and train a dataset
• Creating Lightning Components to recognize an image