Operationalizing and generating value

Operationalizing a model in your infrastructure can be a complicated undertaking. There are some aspects of deployment that are made simple by DataRobot, but there are other parts of deployment that are outside the scope of DataRobot and can be quite challenging. Let's discuss the tasks that are part of this process, as follows:

• Deploying a model as an application programming interface (API): One of the very first tasks is to deploy your model as an API so that it can serve predictions as needed. You will have to decide whether this needs to be a batch or real-time operation. DataRobot automates much of the task of setting this up, and you can have an API serving predictions in minutes.
• Integration and testing with business systems: Having an API is only part of the story—you will now need to integrate this API into your business systems. Sometimes, you can serve up predictions to users via standalone Excel files or web...