Abstract: A method for managing conversation history in an AI system is disclosed. A user message is received and full conversation history payload is retrieved from a memory service. The full conversation history payload includes one or more messages, summaries, or moments. A prompt budget is dynamically allocated based on an available context window. The prompt budget determines token allocations for the messages, summaries, or moments from the full conversation history payload. A prompt is assembled for a language model by selecting conversation history elements to fit within the allocated prompt budget. The selecting balances between recent verbatim messages, summarized content of older messages, and relevant older moments. The prompt budget is iteratively adjusted and the prompt is iteratively reassembled as new information is added from one or more language model outputs or tool calls, while maintaining the full conversation history payload in the memory service.

Abstract: Techniques and systems are described that perform automated identification and retrieval of contextual information for quick and accurate processing of user queries by artificial intelligence generative models. The techniques include receiving a natural language (NL) query associated with a user identifier (ID) and obtaining, using a first NL generative model, contextual data that is pertinent to the NL query and is associated with the user ID. The techniques further include generating an augmented NL query that is based on the NL query and the contextual data. The techniques include communicating the augmented NL query to a recipient that includes the first NL generative model, a second NL generative model, or a user session associated with the user ID.

Abstract: Techniques and systems are described that perform automated identification and retrieval of contextual information for quick and accurate processing of user queries by artificial intelligence generative models. The techniques include receiving a natural language (NL) query associated with a user identifier (ID) and obtaining, using a first NL generative model, contextual data that is pertinent to the NL query and is associated with the user ID. The techniques further include generating an augmented NL query that is based on the NL query and the contextual data. The techniques include communicating the augmented NL query to a recipient that includes the first NL generative model, a second NL generative model, or a user session associated with the user ID.

Abstract: Systems and methods for modifying a user interface of a device based on unavailability of an item offered for sale by a merchant are described. In examples, the techniques include receiving, by server(s) of a service provider, menu data indicating items offered for sale by the merchant. Based on the menu data, instructions that cause presentation of a user interface comprising icons corresponding to the items are sent by the server(s) to device(s) of a merchant and/or customer(s) of the merchant. The techniques further include the server(s) receiving an indication that one of the items is unavailable for sale by the merchant. Responsive to receiving the indication, instructions are sent by the server(s) to at least one of the devices, wherein the instructions cause presentation of a modified user interface that excludes from the user interface an icon corresponding to the unavailable item.

Abstract: Techniques and systems are described that perform automated identification and retrieval of contextual information for quick and accurate processing of user queries by artificial intelligence generative models. The techniques include receiving a natural language (NL) query associated with a user identifier (ID) and obtaining, using a first NL generative model, contextual data that is pertinent to the NL query and is associated with the user ID. The techniques further include generating an augmented NL query that is based on the NL query and the contextual data. The techniques include communicating the augmented NL query to a recipient that includes the first NL generative model, a second NL generative model, or a user session associated with the user ID.

Abstract: Systems and methods for automatically modifying a user interface of a mobile computing device in a restaurant are described. In examples, the techniques include identifying selectable icons to display on the user interface, each of which represents an item offered for sale. The mobile computing device may comprise a first sensor for determining a location of the mobile computing device and a second sensor for determining environmental factors associated with the restaurant. At a first time and based on output of the first sensor, a location of the mobile computing device may be determined, and a set of selectable icons may be presented. Based at least in part on data from the second sensor, a change in environmental factor(s) may be determined. Based on the change in the second context, selectable icons may be dynamically modified and presented. The location and environmental factor(s) are determined without a user input.

Abstract: This disclosure describes an enhanced user interface. For instance, a system may present a user interface indicating a first layout of a physical establishment and receive a first input indicating an object is located at a first geographic location within the physical establishment. The system may receive a recommended location corresponding to a second geographic location within the physical establishment and present the indication of the recommended location for the object on the user interface indicating the first layout. The system may receive image data associated with the physical establishment and determine based at least in part on analyzing the image data that the object is located at the second geographic location within the physical establishment. The system may further display a second layout of the physical establishment, where the second layout includes an indication of the object at the second geographic location within the physical establishment.

Abstract: This disclosure describes an enhanced user interface. For instance, a system may present a user interface indicating a first layout of a physical establishment and receive a first input indicating an object is located at a first geographic location within the physical establishment. The system may receive a recommended location corresponding to a second geographic location within the physical establishment and present the indication of the recommended location for the object on the user interface indicating the first layout. The system may receive image data associated with the physical establishment and determine based at least in part on analyzing the image data that the object is located at the second geographic location within the physical establishment. The system may further display a second layout of the physical establishment, where the second layout includes an indication of the object at the second geographic location within the physical establishment.

Abstract: Described herein is a centralized order computing system that is configured to receive orders from a plurality of sources. The plurality of sources can include delivery applications, remote customer computing devices, and/or service computing devices associated with a merchant. The centralized order computing system may be configured to receive a plurality of orders from the plurality of sources, send the plurality of orders to a kitchen computing system associated with the merchant, receive a notification of order completion, and process the order.

Abstract: Described herein is a centralized order computing system that is configured to receive orders from a plurality of sources. The plurality of sources can include delivery applications, remote customer computing devices, and/or service computing devices associated with a merchant. The centralized order computing system may be configured to receive a plurality of orders from the plurality of sources, send the plurality of orders to a kitchen computing system associated with the merchant, receive a notification of order completion, and process the order.

Abstract: Described herein is a centralized order computing system that is configured to receive orders from a plurality of sources. The plurality of sources can include delivery applications, remote customer computing devices, and/or service computing devices associated with a merchant The centralized order computing system may be configured to receive a plurality of orders from the plurality of sources, send the plurality of orders to a kitchen computing system associated with the merchant, receive a notification of order completion, and process the order.

Abstract: Described herein is a centralized order computing system that is configured to receive orders from a plurality of sources. The plurality of sources can include delivery applications, remote customer computing devices, and/or service computing devices associated with a merchant. The centralized order computing system may be configured to receive a plurality of orders from the plurality of sources, send the plurality of orders to a kitchen computing system associated with the merchant, receive a notification of order completion, and process the order.

Abstract: This disclosure describes, in part, techniques for analyzing sensor data to identify a bottleneck location within a physical establishment, and then providing a recommendation based on the bottleneck location. For instance, a payment service may analyze a layout of a physical establishment using sensor data representing movements of users within the physical establishment. Based on the analysis, the payment service can identify a location within the physical establishment in which movements of the users correspond to one another. For instance, the payment service can identify that the users are located at a similar location within the physical establishment at a given time period. The payment service can then generate and provide a message that indicates a location for moving an object such that users can utilize new routes through the physical establishment.

Abstract: Techniques and arrangements for dynamically modifying a user interface on a computing device of a point-of-sale (POS) system based on a merchant inventory. The user interface can include one or more icons corresponding to selectable items provided by the merchant, such as those available in the merchant inventory. A modification to the user interface can include an adjustment to a visual appearance of an icon and/or a replacement of an icon with another icon based on an analysis of the merchant inventory.

Abstract: This disclosure describes, in part, techniques for analyzing sensor data to identify a bottleneck location within a physical establishment, and then providing a recommendation based on the bottleneck location. For instance, a payment service may analyze a layout of a physical establishment using sensor data representing movements of users within the physical establishment. Based on the analysis, the payment service can identify a location within the physical establishment in which movements of the users correspond to one another. For instance, the payment service can identify that the users are located at a similar location within the physical establishment at a given time period. The payment service can then generate and provide a message that indicates a location for moving an object such that users can utilize new routes through the physical establishment.

Abstract: Techniques and arrangements for dynamically modifying a user interface on a computing device of a point-of-sale (POS) system based on a location of the computing device. The location can be associated with a functionality of the user interface. The user interface can include one or more icons corresponding to selectable items that correspond to the functionality. A modification to the user interface can include an adjustment to a visual appearance of an icon and/or a replacement of an icon with another icon.

Abstract: Techniques and arrangements for dynamically modifying a user interface on a computing device of a point-of-sale (POS) system based on contextual data. The contextual data can include one or more contextual factors surrounding the use of the computing device. The user interface can include one or more icons corresponding to selectable items based on the contextual data. A modification to the user interface can include an adjustment to a visual appearance of an icon and/or a replacement of an icon with another icon.

Abstract: Described herein is a centralized order computing system that is configured to receive orders from a plurality of sources. The plurality of sources can include delivery applications, remote customer computing devices, and/or service computing devices associated with a merchant. The centralized order computing system may be configured to receive a plurality of orders from the plurality of sources, send the plurality of orders to a kitchen computing system associated with the merchant, receive a notification of order completion, and process the order.

Abstract: Techniques and arrangements for dynamically modifying a user interface on a computing device of a point-of-sale (POS) system based on a location of the computing device. The location can be associated with a functionality of the user interface. The user interface can include one or more icons corresponding to selectable items that correspond to the functionality. A modification to the user interface can include an adjustment to a visual appearance of an icon and/or a replacement of an icon with another icon.

Abstract: Techniques and arrangements for dynamically modifying a user interface on a computing device of a point-of-sale (POS) system based on contextual data. The contextual data can include one or more contextual factors surrounding the use of the computing device. The user interface can include one or more icons corresponding to selectable items based on the contextual data. A modification to the user interface can include an adjustment to a visual appearance of an icon and/or a replacement of an icon with another icon.