Abstract: A technical support system identifies a plurality of technical support cases that were previously resolved and identifies related technical support cases of the plurality of technical support cases based on problem descriptions and resolution summaries associated with the plurality of technical support cases. A Large Language Model (LLM) is used to create an initial contact template for potential technical support issues associated with the related technical support cases. Question and answer pairings associated with the potential technical support issues are identified based on communications in the related technical support cases. A machine learning model is trained to communicate with potential users associated with the potential technical support issues based on the question and answer pairings. The initial contact template and the machine learning model are used to resolve a subsequently received technical support issue associated with the related technical support cases.

Abstract: This disclosure describes techniques for diagnosing a presence or malfunction of a network node. In an example method, a first network node receives an indication of a diagnostic transmission originating from a second network node. The second network node further receives a forwarded transmission corresponding to the diagnostic transmission. The first network node diagnoses at least one of a presence or a malfunction of an intermediary node between the first network node and the second network node based on at least one of the indications of the diagnostic transmission or the forwarded transmission.

Abstract: This disclosure describes techniques for providing notifications about events that occur during an online meeting. For instance, a system may establish an online meeting, such as a video conferencing meeting, for users. While in the meeting, a first user may view video of a second user and determine that the second user is experiencing an emergency event. As such, a user device of the first user may receive an input indicating that the emergency event is occurring with the second user. The system may receive the indication from the user device and verify that the emergency event is occurring. Additionally, the system may then send a notification to emergency personnel that indicates at least that the emergency event is occurring and a location of the second user. Furthermore, the system may send notification(s) to contact(s) associated with the second user that indicate at least that the emergency event is occurring.

Abstract: A network management system tests the availability of a network resource before a user performs a task with the network resource. The system measures network activity of a user performing one or more tasks. The network activity includes communication between a user device of the user and each network resource associated with a corresponding task performed by the user. The system also generates a digital model persona of the user based on the tasks performed by the user, and determines a schedule of the tasks performed the user. Each particular task is associated with a corresponding execution time for the user. The system further configures the digital model persona to test the network resource associated with each corresponding task at a testing time that is a predetermined length of time prior to the execution time for the user.

Abstract: Techniques are described to enable selective content sharing in a video conference. In one example, a conferencing system obtains first display content from a first endpoint device in a video conference session. The conferencing system streams the first display content to a second endpoint device in the video conference session and obtains second display content from the second endpoint device. While streaming the first display content to the second endpoint device, the conferencing system streams the second display content to a third endpoint device in the video conference session and prevents streaming of the second display content to the first endpoint device.

Abstract: A method performed at an analysis server, comprising: identifying data objects used by logic configured to produce alternative outcomes depending on first values of the data objects; obfuscating the data objects to produce obfuscated data objects that are related to, but do not identify, the data objects; sending, to a target device, individual requests for second values of the obfuscated data objects; receiving, from the target device, individual responses to the individual requests, wherein the individual responses include obfuscated second values that are related to but do not identify the second values; and evaluating the logic using the obfuscated second values in place of the first values, and without de-obfuscating the obfuscated second values, to produce one of the alternative outcomes.

Abstract: Presented herein are techniques to perform call failure diagnostics. A method includes receiving, at a network device, an indication of calls-of-interest, detecting, at the network device, a failure of one of the calls-of-interest, triggering, in response to the detecting, at the network device, diagnostics data analysis of data associated with the failure of one of the calls-of-interest, determining, based on the diagnostics data analysis, a cause of the failure of the one of the calls-of-interest, and notifying, by the network device, a management system of the cause of the failure of the one of the calls-of-interest and of recent configuration changes on the network device that are related to the cause of the failure of the one of the calls-of-interest.

Abstract: In one embodiment, an issue analysis service obtains telemetry data from a plurality of devices in a network across a plurality of time intervals. The service detects a failure event in which a device in the network is in a failure state. The service clusters the telemetry data obtained prior to the failure event into rounds according to time intervals in which the telemetry data was collected. Each round corresponds to a particular time interval. The service applies a machine learning-based classifier to each one of the rounds of clustered telemetry data to identify one or more common traits appearing in the telemetry data for each round. The service generates a trait change report indicating a change in the one or more common traits appearing in the telemetry data across the rounds leading up to the failure event.

Abstract: A network management station is configured to test a computer network through digital model personas. The network management station obtains network behavior data corresponding to user(s) of the computer network and generates digital model(s) based on the network behavior. The network management station deploys a first digital model persona based on a first digital model among the generated digital models. The first digital model persona operates at a first network location to test the computer network.

Abstract: A computer executed process for mimicking human dialog, referred to herein as a “humanoid” or “humanoid system,” can be configured to provision itself to provide automated customer support. The humanoid can be trained for a customer support campaign. The training can include the humanoid observing communications between a human operator and at least one customer regarding at least one customer support case in the customer support campaign. The humanoid can assess at least one confidence level of the humanoid for the customer support campaign to determine whether the humanoid is adequately trained to handle future customer support cases for the customer support campaign. The humanoid can provision itself to handle at least one future customer support case in the customer support campaign in response to determining that it is adequately trained for the customer support campaign.

Abstract: Techniques for peer-to-peer backup in an administrative domain are described. The techniques may include identifying, by a first device, a second device as a trusted peer device. The techniques may also include receiving an advertisement that the second device supports a peer-to-peer backup system and transmitting an acknowledgement to the second device that the first device supports the peer-to-peer backup system. Additionally, techniques may include receiving a request to store backup data associated with the second device on the first device, the request may include an amount of memory needed. If the first device has the needed memory, the first device may transmit an indication that the memory is available. The indication may include a recovery key. The first device may then receive backup data associated with the second device, encrypted with the recovery key, and store the encrypted backup data in memory reserved for peer-to-peer backup data.

Abstract: A computer executed process for mimicking human dialog, referred to herein as a “humanoid” or “humanoid system,” can be configured to provision itself to provide automated customer support. The humanoid can be trained for a customer support campaign. The training can include the humanoid observing communications between a human operator and at least one customer regarding at least one customer support case in the customer support campaign. The humanoid can assess at least one confidence level of the humanoid for the customer support campaign to determine whether the humanoid is adequately trained to handle future customer support cases for the customer support campaign. The humanoid can provision itself to handle at least one future customer support case in the customer support campaign in response to determining that it is adequately trained for the customer support campaign.

Abstract: The techniques described herein relate to a method including: providing input to a plurality of prediction models; obtaining an initial prediction from each of the plurality of prediction models; providing the input to one or more weight models; obtaining from the one or more weight models a weight for each initial prediction, wherein the weight for each initial prediction is based upon the input and behavior of each of the plurality of prediction models; and determining an output prediction from the initial predictions and the weights.

Abstract: A virtual collaboration system provides communication between a plurality of participants via a collaboration session that includes a corresponding plurality of participant devices. The collaboration session facilitates communication between the participant devices via a plurality of communication channels over which the participants communicate. For example, in some embodiments, the channels include one or more of a video channel, audio channel, or document sharing channel. A participant designates in which of the plurality of channels they will participate, and the designations are communicated to other participants. This provides a broader understanding of each participant's circumstances and ability to engage across the different channels with the plurality of participants.

Abstract: A computer executed process for mimicking human dialog, referred to herein as a “humanoid” or “humanoid system,” can be configured to provide automated customer support. The humanoid can identify a support issue for a customer, as well as a customer support campaign corresponding to the support issue. The humanoid can identify at least one machine learning model associated with the customer support campaign and can communicate with the customer using the at least one machine learning model. The humanoid can execute a support action to resolve the support issue.

Abstract: A computer executed process can be configured to identify a network traffic flow of a source network device. The process can select a traffic fidelity rate from a plurality of traffic fidelity rates for sending information regarding the network traffic flow to a digital twin device corresponding to the source network device. The process can transmit the information regarding the network traffic flow to the digital twin device according to the traffic fidelity rate. The digital twin device can be configured to replicate at least one operation of the source network device with respect to the network traffic flow.

Abstract: Presented herein are techniques to perform call failure diagnostics. A method includes receiving, at a network device, an indication of calls-of-interest, detecting, at the network device, a failure of one of the calls-of-interest, triggering, in response to the detecting, at the network device, diagnostics data analysis of data associated with the failure of one of the calls-of-interest, determining, based on the diagnostics data analysis, a cause of the failure of the one of the calls-of-interest, and notifying, by the network device, a management system of the cause of the failure of the one of the calls-of-interest and of recent configuration changes on the network device that are related to the cause of the failure of the one of the calls-of-interest.

Abstract: This disclosure describes techniques for diagnosing a presence or malfunction of a network node. In an example method, a first network node receives an indication of a diagnostic transmission originating from a second network node. The second network node further receives a forwarded transmission corresponding to the diagnostic transmission. The first network node diagnoses at least one of a presence or a malfunction of an intermediary node between the first network node and the second network node based on at least one of the indications of the diagnostic transmission or the forwarded transmission.

Abstract: Presented herein are techniques associated with proactively scheduling shared resources or responsibilities among multiple users. In one example a method is provided that may include sharing masked calendar information between a first device of a first user and at least one second device of at least one second user based on the first device being proximate to the at least one second device; identifying a conflict involving use of a shared resource at a particular time; comparing first priority metadata for the first user and second priority metadata for the at least one second user; and assigning the use of the shared resource to one of the first user or the at least one second user based, at least in part, on determining that the first user or the at least one second user has a highest priority for use of the shared resource.

Abstract: A network management system tests the availability of a network resource before a user performs a task with the network resource. The system measures network activity of a user performing one or more tasks. The network activity includes communication between a user device of the user and each network resource associated with a corresponding task performed by the user. The system also generates a digital model persona of the user based on the tasks performed by the user, and determines a schedule of the tasks performed the user. Each particular task is associated with a corresponding execution time for the user. The system further configures the digital model persona to test the network resource associated with each corresponding task at a testing time that is a predetermined length of time prior to the execution time for the user.