Model Builder is an easy-to-use AI platform to build, train, and deploy Google’s Vertex AI models leveraging Data Cloud for insights …
Here are highlights from article Bring Your Google Vertex AI Models to Data Cloud
1. Model Builder integration with external AI platforms:
– Model Builder can integrate with external AI platforms such as Google Cloud Vertex AI and Amazon SageMaker.
– This allows users to build, train, and deploy custom AI models externally using data from Salesforce Data Cloud.
2. General availability of Google Vertex AI models in Model Builder:
– Model Builder now supports Google Vertex AI models, in addition to Amazon SageMaker.
– This expands the options for users to leverage AI models in Salesforce.
3. Bring your own model (BYOM) capability:
– Model Builder’s BYOM capability enables users to connect external predictive models to Salesforce.
– Users can use these models to score leads, recommend products, or detect churn across Salesforce workflows.
4. Benefits of using Google Cloud Vertex AI model with Data Cloud in Model Builder:
– Access to highly curated, harmonized, and near real-time data across Customer 360 in Vertex AI.
– Elimination of tedious and error-prone ETL jobs, reducing overhead and improving efficiencies.
– Quick model development and connection with Data Cloud on a single platform.
– Support for real-time, streaming, and batch data ingestion for relevant AI outputs.
– Automation of business processes in Salesforce Data Cloud with Vertex AI predictions.
5. Application workflow for using Model Builder with Google Cloud’s Vertex AI:
– Python connector gives Vertex AI secure access to Salesforce Data Cloud objects.
– Data specialists can explore and prepare data, perform feature engineering tasks, and develop AI models using Vertex AI.
– API gateway may be required for API key-based authentication with Vertex AI endpoint.
6. New features in Model Builder:
– Authentication using Google Service account credentials.
– Streaming ingestion of data.
You can read it here: https://sfdc.blog/nBZnF
Source from developer(dot)salesforce(dot)com