Abstract: Technologies are shown for automatic account generation for a transaction between two user devices. In accordance with some aspects, machine readable code is received at a first user device of a first user from a second user device of a second user. The machine readable code is associated with an item token for a transaction between the users. Responsive to receiving the machine readable code: an authentication token is received in response to authenticating the first user, and an account creation request that includes the item token and the authentication token is sent to an intermediary transaction service. The account creation request causes the intermediary transaction service: to generate a first account for the first user using information for the first user obtained using the authentication token, and to facilitate a transaction for the item between the first account and a second account of the second user.

Abstract: Technologies are shown for buyer initiated automatic seller account generation where a buyer client receives an identification of a purchase item, generates an item token corresponding to the purchase item, and provides the item token to a seller client. The seller client authenticates a seller user with an authentication service, such as a biometric authentication service, receives an authentication token corresponding to the seller user, and sends an account creation request with the item token and the authentication token to an account creation service. The account creation service receives the account creation request, sends a user information request to the authentication service with the authentication token and receives seller information for the seller user from the authentication service. The account creation service creates a seller user account using the seller information, commits a purchase transaction for the purchase item, and sends an item purchase confirmation to the buyer client.

Abstract: Technologies are shown for artificial intelligence agents for predicting items of interest utilizing multiple data sources, such as historical user behavior, item wear profiles, inventory data and social network data. User models to the multiple sources of data to predict an item of interest to the user. Search requests pertaining to the predicted item can be generated and submitted to electronic commerce platforms and results responsive to the first set of search requests pertaining to the first predicted item received. In one aspect, one or more of the search results can be selected for display to the user. A search result selected by the user can be received and a purchase transaction committed. In another aspect, the agent is authorized to autonomously execute a purchase transaction on a selected one of the search results. Different model types can be utilized for predicting different types of items.

Abstract: Technologies are shown for buyer initiated automatic seller account generation where a buyer client receives an identification of a purchase item, generates an item token corresponding to the purchase item, and provides the item token to a seller client. The seller client authenticates a seller user with an authentication service, such as a biometric authentication service, receives an authentication token corresponding to the seller user, and sends an account creation request with the item token and the authentication token to an account creation service. The account creation service receives the account creation request, sends a user information request to the authentication service with the authentication token and receives seller information for the seller user from the authentication service. The account creation service creates a seller user account using the seller information, commits a purchase transaction for the purchase item, and sends an item purchase confirmation to the buyer client.

Abstract: Example embodiments provide a prediction-based caching system. The caching system receives a request for current requested information from a client device. The caching system causes retrieval of the current requested information from one of a cache or a source. The caching system then transmits the current requested information to the client device. The caching system also determines a predicted request for information. The caching system retrieves predicted results based on the predicted request from the source, and stores the predicted results in a cache for faster retrieval.

Abstract: Example embodiments provide a prediction-based caching system. The caching system receives an application program interface (API) request from a component of the client device. The request indicates a parameter for current requested information. The caching system causes a search of both a cache and a source for the current requested information based on the parameter. The caching system causes retrieval of the current requested information from one of the cache or the source. The caching system then transmits the current requested information to the component of the client device. The caching system also predicts a future request for information (or receives a predicted request for future information). The caching system retrieves predicted results based on the predicted request from the source, and stores the predicted results in a cache for faster retrieval.

Abstract: Example embodiments provide a system and method for providing prediction-based instant search results. A system detects a current input entered in a search field of a user interface by a user. The system generates an auto-suggestion application program interface (API) request that includes the current input as a parameter, and transmits the auto-suggestion API request to an auto-suggestion server. In response, the system receives auto-suggestions from the auto-suggestion server. The system generates a search application program interface (API) request that includes the current input and at least some of the auto-suggestions as parameters, and transmits, the search API request to a search system. In response, the system receives results. The system then updates the user interface to display the results.

Abstract: Example embodiments provide a prediction-based caching system. The caching system receives an application program interface (API) request from a component of the client device. The request indicates a parameter for current requested information. The caching system causes a search of both a cache and a source for the current requested information based on the parameter. The caching system causes retrieval of the current requested information from one of the cache or the source. The caching system then transmits the current requested information to the component of the client device. The caching system also predicts a future request for information (or receives a predicted request for future information). The caching system retrieves predicted results based on the predicted request from the source, and stores the predicted results in a cache for faster retrieval.