Einstein Studio is a new technology from Salesforce that makes it easy for businesses to use their proprietary (owned data) to build and deploy AI models.
Salesforce follows an agnostic approach to large language models (LLMs). You can use Salesforce’s own LLM, bring your own model (BYOM) from other platforms (e.g. Amazon SageMaker, Google Vertex AI), and combine your company data with preferred AI models from those other predictive or generative AI services.
Why is this a big deal? Organizations are rapidly learning how they can adopt AI – driven by piqued interest from the C-Suite and with the goal to remain competitive – and how they can do so responsibly to respect data privacy.
In this rush, there’s been a mismatch between excitement and reality. In one survey, approximately 60% of executives said their organization is 1-2 years away from implementation, with 54% of AI projects failing to make it from pilot to production.
What Is Einstein Studio?
Einstein Studio enables you to combine your company data with preferred AI models from those other predictive or generative AI services. This is an additional option, blending data privacy (your organization owns its data) with the power of AI models from other services.
Data science and engineering teams can easily build, train, and activate custom AI models using a bring your own model (BYOM) approach within Salesforce, which:
- Means you can build and train custom AI models to use across Salesforce applications to inform predictions and auto-generated content.
- Removes the manual and time-intensive burdens associated with AI integrations and model training.
- Uses a zero-ETL (extract, transform, and load) framework, to infuse advanced AI models with the most relevant data across sales, service, marketing, commerce, and IT. Get more from existing data – there’s no integration required.
How Einstein Studio Works
Einstein Studio is an interface that guides you through each step of this unique infusion process between your custom models (built externally) and your CRM data.
The Model Builder prompts you to choose your type of model, which could be to:
- Create a model from scratch
- Connect a SageMaker model
- Connect a Google Vertex AI model
- Connect a model on another platform
In the Welcome to Model Builder side panel you’ll see further guidance on each option, such as: Which model type you should use, how to create one from scratch, and how to register a model.
Once the model is activated, you can see the details on its own record page. This includes Prediction Jobs which keep a log of the model’s usage within your CRM. Here you will see useful information including:
- Job Name
- Type
- Status
- Last Run Date
- Last Run Status
Einstein Studio Models Are Extensible
One driving ethos of the Salesforce platform is that items created in one place on the platform can be utilized in other areas – much like building blocks that can be reused to bring additional capabilities.
Models developed in Einstein Studio can be leveraged in the Salesforce Flow Builder to inject GenAI recommendations into automated processes – pictured below, we can see that the flow is set to evaluate the output from the “check prediction” step, and send email subscribers down one path to receive the product recommendation email or the alternative path.
Summary
Salesforce have stepped up to the mark to introduce GenAI CRM (the next generation of CRM) to our organizations – it’s certainly an exciting time to be a Salesforce professional with the opportunities that have appeared in a matter of months.
Einstein Studio solves one of the tech industry’s biggest AI adoption challenges by allowing you to easily combine your company data with preferred AI models from other predictive or generative AI services. This plays a key part in equipping your users to turn AI into action.
While Salesforce has developed their own LLM, there are other popular choices on the market (Amazon SageMaker and Google Vertex AI, to name just a couple) that have been adopted by technical teams globally, Salesforce advocate the open ecosystem approach, giving your organization the freedom to choose the providers of their teams’ choice.
This is very interesting news for the Salesforce ecosystem, and we should continue to see Salesforce provide us with the tools – like the Einstein Studio interface – to make GenAI a powerful addition to our organizations.
Read more:
- Salesforce and AI – Your Long-Term Career Path
- AI Wars: How Salesforce’s Agnostic LLM Approach Works
- What the C-Suite is Thinking About Salesforce + AI