Improved custom model support allows you to make workflow decisions based on a propriety data set or model
Alloy now allows you to make workflow decisions based on a proprietary dataset or model. For example, your company may have its own fraud model based on your own device and location data that you would like to apply in a workflow. Custom models were previously supported but now clients can set up these models much more quickly and independently.
We now allow clients to configure a “Custom Model” service within an Alloy workflow in the following way:
- You host and configure an endpoint that Alloy will POST to during an evaluation with any service data that have been configured to run before the custom model
- Your model returns a score to Alloy
- You can configure decision thresholds based on that score in the Alloy dashboard
To learn more about how to use custom models, please reference our support documentation.