Automated agent messaging system

1. A method for customizing electronic messages, the method comprising:

receiving, via a messaging application of a computing system and from an agent of an enterprise, an electronic message that (i) includes textual content intended for delivery to a customer of the enterprise, and (ii) is to be delivered to an electronic device operated by the customer;

revising, via the message application, the electronic message while maintaining an intent of the textual content of the electronic message, the revising including:

applying one or more machine learning models to the textual content of the electronic message and to an indication of items included in an order of the customer with the enterprise to thereby:

replace one or more portions of the textual content of the electronic message with predefined textual content associated with the intent;

select, from among a group of items included in the order of the customer, a particular item that has a higher degree of association to occurrences of subsequent customer activities as compared to respective degrees of association of other items, of the group of items, to the occurrences of the subsequent customer activities; and

determine a type of a most likely subsequent customer activity associated with the particular item;

adding, to the electronic message, additional textual content that is separate from any intent corresponding to the one or more portions, the additional textual content corresponding to the particular item selected from among the group of items included in the order of the customer and the type of the most likely customer activity; and

transmitting, by the computing system, the revised electronic message to the electronic device operated by the customer.

2. The method of claim 1 , wherein applying the one or more machine-learning models to the textual content of the electronic message comprises applying a natural language understanding (NLU) model to the textual content of the electronic message, thereby determining the intent of the textual content of the electronic message.

3. The method of claim 1 , wherein applying the one or more machine learning models to the textual content of the electronic message and to the indication of items included in the order of the customer includes applying the one or more machine learning models to the textual content of the electronic message, the indication of items included in the order of the customer, and historical data specific to the customer.

4. The method of claim 1 , further comprising:

receiving, by the computing system, feedback for the revised electronic message; and

updating the one or more machine learning models based on the feedback.

5. The method of claim 1 , wherein revising the electronic message further comprises:

selecting, from a plurality of predefined messages that are indexed to a plurality of respective intents, a particular predefined message, the particular predefined message indexed to the intent of the textual content of the electronic message; and

including at least a portion of the particular predefined message in the predefined textual content.

6. The method of claim 1 , wherein applying the one or more machine learning models includes:

applying at least one machine learning model that has been trained based on a machine analysis of at least one of historical data associated with the enterprise or historical data specific to the customer; and

generating the additional textual content based on an output of the at least one machine learning model.

7. The method of claim 1 , wherein revising the electronic message further comprises:

determining that the intent of the textual content of the electronic message is omitted from a plurality of respective intents to which a plurality of predefined messages are indexed;

based on the omission, defining a new message and including the defined new message in the revised electronic message, the new message corresponding to the intent of the textual content of the electronic message; and

updating the plurality of predefined messages to include the new message.

8. The method of claim 1 , wherein:

the determination of the type of the most likely subsequent customer activity associated with the particular item includes a determination of one or more subsequent customer activities that have respective higher degrees of association to the selected particular item as compared to respective degrees of association of other subsequent customer activities to the selected particular item; and

the additional textual content further corresponds to the determined one or more subsequent customer activities.

9. A computing system for customizing electronic messages, the computing system comprising:

one or more memories storing a messaging application; and

one or more processors coupled to the one or more memories, the one or more processors configured to execute the message application to:

receive, from an agent of an enterprise, an electronic message that (i) includes textual content intended for delivery to a customer of the enterprise, and (ii) is to be delivered to an electronic device operated by the customer;

revise the electronic message while maintaining an intent of textual content of the electronic message, the revision including an application of one or more machine learning models to the textual content of the electronic message and to an indication of items included in an order of the customer with the enterprise to thereby:

replace one or more portions of the textual content of the electronic message with predefined textual content associated with the intent;

select, from among a group of items included in the order of the customer, a particular item that has a higher degree of association to occurrences of subsequent customer activities as compared to respective degrees of association of other items, of the group of items, to the occurrences of the subsequent customer activities; and

determine a type of a most likely subsequent customer activity associated with the particular item;

add, to the electronic message, additional textual content that is separate from any intent corresponding to the one or more portions, the additional textual content corresponding to the particular item selected from among the group of items included in the order of the customer and the type of the most likely customer activity; and

cause the revised electronic message to be transmitted to the electronic device operated by the customer.

10. The computing system of claim 9 , wherein:

the one or more machine learning models include a natural language understanding (NLU) model; and

the messaging application is executable to apply the NLU model to the textual content of the electronic message to thereby determine the intent of the textual content of the electronic message.

11. The computing system of claim 9 , wherein the messaging application applies the one or more machine learning models to the textual content of the electronic message and to the indication of the items included in the order of the customer in conjunction with historical data specific to the customer.

12. The computing system of claim 9 , wherein the computing system is configured to receive feedback for the revised electronic message and update the one or more machine learning models based on the feedback.

13. The computing system of claim 9 , wherein:

the messaging application is further executable to select, from a plurality of predefined messages that are indexed to a plurality of respective intents, a particular predefined message, the particular predefined message indexed to the intent of the textual content of the electronic message; and

at least a portion of the particular predefined message is included in the predefined textual content.

14. The computing system of claim 9 , wherein:

at least one of the one or more machine learning models is trained based on a machine analysis of at least one of historical data specific to the customer or historical data associated with the enterprise;

the messaging application is further executable to apply the at least one of the one or more machine learning models to information corresponding to the customer to thereby generate an output of the at least one of the one or more machine learning models; and

the additional textual content is based on the output of the at least one of the one or more machine learning models.

15. The computing system of claim 14 , wherein the output of the at least one of the one or more machine learning models is indicative of at least one of:

the selected particular item or the determined type of the most likely subsequent customer activity.

16. The computing system of claim 9 , wherein the messaging application is further executable to:

determine that the intent of the textual content of the electronic message is omitted from a plurality of respective intents to which a plurality of predefined messages are indexed; and

based on the omission, include a new message in the revised electronic message, the new message corresponding to the intent of the textual content of the electronic message.

17. The computing system of claim 9 , wherein:

the messaging application is further executable to:

cause the electronic message and the revised electronic message to be displayed on a user interface;

obtain a selection of the revised electronic message responsive to the display of the electronic message and the revised electronic message; and

the transmission of the revised electronic message is responsive to the selection.

18. The computing system of claim 9 , wherein:

the determination of the type of the most likely subsequent customer activity associated with the particular item includes a determination of one or more subsequent customer activities that have respective higher degrees of association to the selected particular item as compared to respective degrees of association of other subsequent customer activities to the selected particular item; and

the additional textual content further corresponds to the determined one or more subsequent customer activities.