People.ai, Inc.

Devin Rice

Oleg Rogynskyy

Yurii Brunets

Yury Markovsky

Stefan Hermanek

Jose Alberto Muniz Navarro

Wei Hai

James Wilson

Rajit Kurien Joseph

John Wulf

Vardhman Jain

Tetiana Lutsaievska

Kavita Shah

Adam Draper

The present disclosure relates to maintaining an electronic activity derived member node network. Exemplary implementations may: generate a plurality of node profiles; generate a node graph generated from the plurality of node profiles; identify a second electronic activity identifying a pair of electronic accounts; identify using the node graph, a pair of participant node profiles corresponding to the pair of electronic accounts identified by the second electronic activity; identify, using the node graph, an edge between a pair of nodes corresponding to the pair of participant node profiles; determine a relevancy score for the second electronic activity corresponding to the pair of the participant node profiles based on context of the second electronic activity; and update the node graph by modifying the connection corresponding to the edge between the pair of participant node profiles based on the relevancy score determined for the second electronic activity.

Systems and methods for maintaining an electronic activity derived member node network

The present disclosure relates to generating a master group node graph from a plurality of systems of record. Record objects of a system of record corresponding to a data source provider may be accessed. Each record object can have a record object type and identify a group entity. Each record object can be linked to one or more second record objects of a second record object type and identifying a respective member entity. A record object corresponding to a group entity can be identified. A connection profile between the record objects of the group entity and another group entity can be generated. An association between the group node profiles and the connection profile can be stored.

Systems and methods for generating a master group node graph from systems of record

The present disclosure relates to determining the shareability of values of node profiles. Record objects and electronic activities of a system of record corresponding to a data source provider may be accessed. Each record object may correspond to a record object type and have one or more object field-value pairs. Node profiles may be maintained. Values of fields corresponding to a predetermined type of field including fewer than a predetermined threshold number of data source providers may be identified. A restriction tag used to restrict populating other node profiles may be generated. Provision of the value with a second data source provider may be restricted.

Systems and methods for determining the shareability of values of node profiles

The present disclosure relates to methods, systems, and storage media for identifying node hierarchies and connections using electronic activities. The method may include maintaining node profiles corresponding to unique entities; selecting electronic activities linked to both a first node profile and a second node profile of the plurality of node profiles; generating for each electronic activity of the plurality of electronic activities, a relevancy score based on a respective time at which the electronic activity was transmitted or received; generating a connection profile for a connection between the first node profile and the second node profile based on the respective relevancy score of each electronic activity and storing in a data structure, an association between the connection profile, the first node profile, and the second node profile.

Systems and methods for identifying node hierarchies and connections using electronic activities

This disclosure relates to methods, systems, and storage media for generating a performance profile of a node profile. First electronic activities linked to a first node profile can be accessed. For each first electronic activity, a type of the first electronic activity can be determined. A feature extraction policy can be selected to generate a first feature array for the first electronic activity. The first feature array for the first electronic activity can be generated based on the type of the first electronic activity. A first performance profile of the first node profile can be generated for the time period by providing the generated first feature array for each first electronic activity to one or more models trained using second feature arrays corresponding to second electronic activities of second node profiles. An association between the first node profile and the first performance profile can be stored.

Systems and methods for generating a performance profile of a node profile including field-value pairs using electronic activities

The present disclosure relates to generating field-specific health scores for a system of record. Record objects of a system of record corresponding to a data source provider may be accessed. Each record object may correspond to a record object type and have one or more object field-value pairs. Node profiles may be maintained. Each node profile may include one or more node field-value pairs. A subset of node field-value pairs of the node profiles with confidence scores greater than a threshold score may be identified. Node profiles having the subset of node field-value pairs may be identified. Node field-value pairs of the subset may be compared with corresponding object field-value pairs of the record objects. A field-specific health score for each field may be generated based on the comparison.

Systems and methods for generating field-specific health scores for a system of record

The systems and methods described herein can maintain time series of data points. The system can access at least one of a plurality of electronic activities transmitted or received via electronic accounts of one or more data source providers or a plurality of record objects of one or more systems of record of the one or more data source providers; maintain an association between each electronic activity or record object and a timestamp corresponding to when the electronic activity was sent or received by an or when the record object was last updated; maintain a plurality of node profiles; and generate, for a particular time, a confidence score of a value of a field-value pair of a node profile only using the data of the plurality of electronic activities or the plurality of record objects that correspond to the value and have a corresponding timestamp prior to or contemporaneous with the particular time.

Systems and methods for maintaining a time series of data points

The systems and methods described herein can use multiple fields with dynamically determined priorities based on field values for node resolution. The system can generate activity field-value pairs including an activity value associated with an activity field from an electronic activity. The system can determine a frequency score based on a first count of node field-value pairs that match the activity value. The system can assign a weight to the activity value based on the frequency score. The system can generate a match score of a candidate node profile indicating a likelihood that the electronic activity is transmitted or received by an account corresponding to the candidate node profile. The system can store an association between the electronic activity and the node profile selected based on the match score.

Systems and methods for node resolution using multiple fields with dynamically determined priorities based on field values

The present disclosure relates to methods, systems, and storage media for updating field-value pairs of record objects using electronic activities linked to the record objects. The method can maintain a plurality of node profiles corresponding to a plurality of unique entities; access a plurality of electronic activities transmitted or received via electronic accounts associated with one or more data source providers; identify a record object of a plurality of record objects of one or more systems of record; match electronic activity of the plurality of electronic activities to the record object based on content of the electronic activity and the object field-value pairs of the record object; identify a node profile matched with the electronic activity of the plurality of electronic activities; determine a weighting factor for the electronic activity; and update a stage field of the record object to a second stage value indicating a second proximity to the completion of the event.

Systems and methods for updating field-value pairs of record objects using electronic activities

The present disclosure relates to systems and methods for assigning tags to values of node profiles based on detected electronic activity. Exemplary implementations may: maintain node profiles; access a response to an electronic activity transmitted or received via an electronic account; detect responsive to parsing the response, that the response is one of a bounce-back electronic activity indicating that the recipient of the electronic activity is invalid or an automatic responder electronic activity generated by an autoresponder of the recipient; compare a communication identifier of the response to the node profile(s) to identify a node profile having a matching communication identifier; assign a respective tag to the identified at least one node profile responsive to detecting that the electronic activity is the bounce-back electronic activity or the automatic responder electronic activity; and update a node field-value pair of the identified at least one node profile based on the assigned tag.

Growing community of inventors

San Francisco, CA, United States of America

Devin Rice