Abstract: Described herein are methods, systems, and media for knowledge article prediction from historical or open service tickets comprising service ticket summarization and intent extraction and knowledge article prediction in off-line or real-time pipelines.

Abstract: Described herein are methods, systems, and media for ticket field values prediction from historical or open service tickets comprising service ticket summarization and intent extraction and ticket field values prediction in off-line training or real-time prediction pipelines.

Abstract: Described herein are methods and systems for optimizing the retrieval of data from multiple different destination software applications or systems comprising a planner module, a navigator module, an executor module, and a response generator module, which together divide natural language queries of users into manageable sub-tasks using a large language learning model (LLM), convert sub-tasks into computer executable sub-tasks involving LLM-generated code, extract relevant data from destination software systems, and return a natural language result to the user utilizing an LLM. The methods and systems also include internal verifications and user feedback-implemented output modifications.

Abstract: Described herein are methods, systems, and media for deriving operational insights about customer infrastructure comprising (a) obtaining customer incident data reported by one or more customers, wherein the customer incident data is pre-encoded using one or more encoding techniques that are specific or unique to the one or more customers; and (b) using a decoder module to directly process the customer incident data to generate an output, wherein the output comprises a plurality of configuration item(s) (CIs) discovered, derived or extracted from the customer incident data.

Abstract: Described herein are methods, systems, and media for supporting multi-language user sessions and fulfillments comprising: maintaining a repository of fulfillment objects each comprising a language and a region; establishing a user session with a user; determining a user region for the user in association with establishing the user session; identifying one or more fulfillment objects in the repository available for the user region; processing the user session, the user session comprising one or more user requests; applying a language detection model to each request to determine a user request spoken language; applying an understanding module to each request to recommend one or more of the fulfillment objects matching the region for the user; and rendering a response to each request to the user, utilizing the one or more of the fulfillment objects matching the region for the user, in the user request spoken language.

Abstract: A method of selecting and interacting with agent types for an agentic AI system includes deploying an agent based on a template, with the agent assigned to a functional domain required by an enterprise, wherein the agent is selected to be one of a Prescriptive Conversational Agent; a Prescriptive Action Agent; a Q&A Retrieval Agent for private knowledge bases; a Q&A Retrieval Agent for public knowledge bases; a Dynamic Multi-task Action Agent; and a Personal Assistant Agent. The selected agents can be connected to a reasoning engine and orchestration module able to send and receive requests to the agent.

Abstract: An agentic AI system architecture able to respond to a user request includes multiple agents, each agent being assigned to a functional domain required by an enterprise. The agentic AI system architecture also includes at least one reasoning engine supported by at least one large language model (LLM), with the reasoning engine being able to interpret a user request and decompose the user request into one or more mini-tasks. An orchestrator module is connected to the reasoning engine and able to route those mini-tasks based on the user request to at least one of the multiple agents for further processing. Together, the orchestrator module and reasoning engine can collect and reconcile processed results from the multiple agents and prepare a response to the user request.

Abstract: Described herein are methods and systems for optimizing the retrieval of data from multiple different destination software applications or systems comprising a planner module, a navigator module, an executor module, and a response generator module, which together divide natural language queries of users into manageable sub-tasks using a large language learning model (LLM), convert sub-tasks into computer executable sub-tasks involving LLM-generated code, extract relevant data from destination software systems, and return a natural language result to the user utilizing an LLM. The methods and systems also include internal verifications and user feedback-implemented output modifications.

Abstract: In one example embodiment, a server that is in communication with a network that includes a plurality of network elements obtains, from the network, a service request record that includes sensitive information related to at least one of the plurality of network elements. The server parses the service request record to determine that the service request record includes a sequence of characters that is repeated in the service request record, and tags the sequence of characters as a particular sensitive information type. Based on the tagging, the server identically replaces the sequence of characters so as to preserve an internal consistency of the service request record. After identically replacing the sequence of characters, the server publishes the service request record for analysis without revealing the sequence of characters.

Abstract: In one example embodiment, a server that is in communication with a network that includes a plurality of network elements obtains, from the network, a service request record that includes sensitive information related to at least one of the plurality of network elements. The server parses the service request record to determine that the service request record includes a sequence of characters that is repeated in the service request record, and tags the sequence of characters as a particular sensitive information type. Based on the tagging, the server identically replaces the sequence of characters so as to preserve an internal consistency of the service request record. After identically replacing the sequence of characters, the server publishes the service request record for analysis without revealing the sequence of characters.

Abstract: In one example embodiment, a server that is in communication with a network that includes a plurality of network elements obtains, from the network, a service request record that includes sensitive information related to at least one of the plurality of network elements. The server parses the service request record to determine that the service request record includes a sequence of characters that is repeated in the service request record, and tags the sequence of characters as a particular sensitive information type. Based on the tagging, the server identically replaces the sequence of characters so as to preserve an internal consistency of the service request record. After identically replacing the sequence of characters, the server publishes the service request record for analysis without revealing the sequence of characters.

Abstract: In one example embodiment, a server that is in communication with a network that includes a plurality of network elements obtains, from the network, a service request record that includes sensitive information related to at least one of the plurality of network elements. The server parses the service request record to determine that the service request record includes a sequence of characters that is repeated in the service request record, and tags the sequence of characters as a particular sensitive information type. Based on the tagging, the server identically replaces the sequence of characters so as to preserve an internal consistency of the service request record. After identically replacing the sequence of characters, the server publishes the service request record for analysis without revealing the sequence of characters.

Abstract: This disclosure describes techniques for providing a network diagnostic system with on-premise node processing and cloud node processing to optimize bandwidth usage and decrease memory footprint. The on-premise node may receive streaming telemetry from connected network devices and encode to the telemetry data into filtered data objects. The on-premise node may determine whether the state of a network device has changed to determine to push the filtered data object to a cloud node for further diagnostic analysis. The cloud node may include a gateway and a pool of proxy servers, wherein each proxy server is designated to perform diagnostic analysis on a single product type.

Abstract: Techniques for receiving operational preferences for operating network devices, and determining software updates for the network devices based on the operational preferences. A recommendation system may determine a group of network devices in a device network based on the network devices in the group performing a common functional role or have common attributes. The recommendation engine may further receive the operational preferences for the group of network devices from a user associated with the device network. These operational preferences may be continuously, or periodically, evaluated against actual operating conditions of the group of network devices to determine whether a risk metric associated with the actual operation conditions violates an operational preference. In some instances, the recommendation system may provide the user with access to a recommendation to run updated software that is more optimized for the network device and that satisfies the operational preferences of the user.

Abstract: This disclosure describes techniques for providing a network diagnostic system with on-premise node processing and cloud node processing to optimize bandwidth usage and decrease memory footprint. The on-premise node may receive streaming telemetry from connected network devices and encode to the telemetry data into filtered data objects. The on-premise node may determine whether the state of a network device has changed to determine to push the filtered data object to a cloud node for further diagnostic analysis. The cloud node may include a gateway and a pool of proxy servers, wherein each proxy server is designated to perform diagnostic analysis on a single product type.

Abstract: Sequences of computer network log entries indicative of a cause of an event described in a first type of entry are identified by training a long short-term memory (LSTM) neural network to detect computer network log entries of a first type. The network is characterized by a plurality of ordered cells Fi=(xi, ci-1, hi-1) and a final sigmoid layer characterized by a weight vector wT. A sequence of log entries xi is received. An hi for each entry is determined using the trained Fi. A value of gating function Gi(hi, hi-1)=II (wT(hi?hi-1)+b) is determined for each entry. II is an indicator function, b is a bias parameter. A sub-sequence of xi corresponding to Gi(hi, hi-1)=1 is output as a sequence of entries indicative of a cause of an event described in a log entry of the first type.

Abstract: This disclosure describes techniques for providing a network diagnostic system with on-premise node processing and cloud node processing to optimize bandwidth usage and decrease memory footprint. The on-premise node may receive streaming telemetry from connected network devices and encode to the telemetry data into filtered data objects. The on-premise node may determine whether the state of a network device has changed to determine to push the filtered data object to a cloud node for further diagnostic analysis. The cloud node may include a gateway and a pool of proxy servers, wherein each proxy server is designated to perform diagnostic analysis on a single product type.

Abstract: A Services Delivery Platform (SDP) architecture is provided that is configured to onboard new data sets into an SDP data lake. The SDP enables the crowdsourcing of data on-boarding by configuring this process into an interactive, intuitive, step-by-step guided workflow while governing/controlling key functions like verification, acceptance and execution.

Abstract: This disclosure describes techniques for providing a network diagnostic system with on-premise node processing and cloud node processing to optimize bandwidth usage and decrease memory footprint. The on-premise node may receive streaming telemetry from connected network devices and encode to the telemetry data into filtered data objects. The on-premise node may determine whether the state of a network device has changed to determine to push the filtered data object to a cloud node for further diagnostic analysis. The cloud node may include a gateway and a pool of proxy servers, wherein each proxy server is designated to perform diagnostic analysis on a single product type.

Abstract: Presented herein are methods associated with a Services Delivery Platform (SDP) architecture for a distributed application building blocks, such as microservices, deployment-agnostic. The system includes a central compute node and numerous remote compute nodes. Techniques are provided to “onboard” and assimilate the capabilities of remote compute nodes so that they are an integrated part of the SDP system and can be accessed and used in connection with one or more services provided by the SDP system.