Communication routing based on user characteristics and behavior

A routing system configured to perform steps comprising:

• receiving, from a client device of a user, a request to interact with a customer service system;
• generating a feature vector based on an activity history of the client device and a user profile associated with the user;
• inputting the feature vector into a machine learning model configured to map a service provider to entries of the feature vector;
• selecting the service provider based on an output of the machine learning model;
• and
• establishing a communication session between at least the client device and a device of the service provider to fulfill the request to interact.

A non-transitory computer readable storage medium storing executable instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising:

• receiving, from a client device of a user, a request to interact with a customer service system;
• generating a feature vector based on an activity history of the client device and a user profile associated with the user;
• inputting the feature vector into a machine learning model configured to map a service provider to entries of the feature vector;
• selecting the service provider based on an output of the machine learning model;
• and
• establishing a communication session between at least the client device and a device of the service provider to fulfill the request to interact.