Return to site

Personalized Einstein Vision demo in 5 minutes?

Create your automatic Image Classification based on your (or your customer) pictures in 5 minutes with SharinPix App form the AppExchange

An Einstein Vision demo in 5 minutes ??? For Real?

Let's face it, we are all very excited about AI in Salesforce and especially around Einstein Vision (and now Einstein sentiment and intent analysis capabilities as well).


But unfortunately, it takes at least long hours or even days of development to start working with it. And finally validate how it will react for your use case or to prepare a customer demo.


That's exactly why we are coming with this new SharinPix App empower by Einstein Vision!

How it works?

After SharinPix App installation, you will have to setup your Einstein Vision account to be ready to go.

Then you will need only 3 quick steps to check how it works :

  • Choose the objects you want to recognize and find at least 10 pictures per object (100 pictures is a good ratio to start with)
  • Create an Einstein Vision project in Salesforce using SharinPix, enter labels corresponding to the objects you want to recognize, upload the corresponding images and tag them with the corresponding label
  • Save as a dataset, Create a model from a dataset in a click and you are ready to go!

Get SharinPix support for your demo!

On top of this incredible App, If you have any difficulty in setting up a demo, just join us! 

We will dedicate one of our image expert to help you understand what is not working and what to change in your Einstein Vision project to make it reach your goal!

Maintain your Einstein Vision model on the go...

Once ready you may want to use this Einstein Vision in production. You can then rely on a personalized development or on SharinPix auto-tagging feature based on what Einstein Vision can recognize at upload.

But you may have users detecting pictures which are not well recognized by the system.

If so, using SharinPix app, you'll be able to train Einstein Vision to enhance the recognition of this kind of images by uploading several images that looks like the none working one.

Don't use the original image provided to explain the error as you may need it to check if the correction made is sufficient.


This process could be achieved in seconds, uploading more pictures to your project, you just have to tag them and create a new dataset and a new model in few clicks to check if the none working image give better results.

Where to start?

That's the easiest part :

Install SharinPix App from the AppExchange :

And read the complete SharinPix documentation :

And if you want to learn more about the API provided by Salesforce and used behind our product check