Use Model Builder to Integrate Google Vertex AI Models with Salesforce

— by

Model Builder is an easy-to-use AI platform that enables data science and engineering teams to build, train, and deploy AI models …

Here are highlights from article Use Model Builder to Integrate Google Vertex AI Models with Salesforce

1. Model Builder and its capabilities
– Model Builder is a user-friendly platform in Einstein Copilot Studio that allows users to create and operationalize AI in Salesforce.
– It can build, train, and deploy custom AI models using data in Salesforce.
– It integrates with other AI platforms like Amazon SageMaker and Google Cloud Vertex AI.

2. BYOM capabilities and zero-copy approach
– Model Builder’s BYOM (Bring Your Own Model) capabilities allow data specialists to build and deploy custom AI models in Salesforce Data Cloud.
– The models are trained externally, such as in Vertex AI, using data from Data Cloud.
– The zero-copy approach ensures that data from Data Cloud is used to build and train the models in Vertex AI.

3. Predictions and insights for business processes
– Once models are deployed in Salesforce, predictions are automatically and continuously updated in near real-time.
– These predictions can be embedded into business processes and used by business users.
– For example, marketing analysts can create segments and personalize user experiences using the model’s predictions.

4. Key benefits of Model Builder
– Model Builder supports diverse AI and ML use cases across Customer 360, such as customer segmentation, personalization, lead conversion, and case classification.
– It uses familiar tools like TensorFlow, PyTorch, and XGBoost to build and train models.
– It provides AI-based insights that can be integrated into Salesforce workflows to optimize business processes.

5. Architectural overview and operationalizing a product recommendation model
– Data from diverse sources is consolidated and prepared using Data Cloud’s lakehouse technology.
– The training dataset is then used in Vertex AI to train and build the AI models.
– An endpoint is created for model deployment and scoring in Data Cloud.
– Salesforce’s automation flow functionality can be used to integrate the model’s predictions into business processes, such as creating tasks or personalized marketing journeys.

You can read it here: https://sfdc.blog/bzTFG

Source from developer(dot)salesforce(dot)com

Newsletter

My latest updates in your e-mail.