Abstract: In one implementation, a device uses a large language model-based agent to perform a task in a network, to obtain a first result. The device executes code in the network to perform the task, to obtain a second result. The device makes a comparison between the first result and the second result. The device provides, based in part on the comparison, one or more performance metrics for the large language model-based agent for display.

Abstract: In one implementation, a device receives an input request for a large language model-based network troubleshooting agent regarding an issue in a network. The large language model-based network troubleshooting agent performs a lookup of a recipe based on the input request, wherein the recipe comprises contextual information for the issue. The device generates, by the large language model-based network troubleshooting agent, a prompt for a large language model based on the input request and on the recipe. The device provides, by the large language model-based network troubleshooting agent, the prompt to the large language model to troubleshoot the issue in the network.

Abstract: In one implementation, a device receives an input request for a large language model-based troubleshooting agent for a network. The device selects an optimization criterion for the large language model-based troubleshooting agent based on the input request. The device provides the optimization criterion to the large language model-based troubleshooting agent to cause the large language model-based troubleshooting agent to select a particular large language model to process the input request based on the optimization criterion. The device sends, to a user interface, an indication of a result of the particular large language model processing the input request.

Abstract: In one implementation, a device performs, using a using a large language model-based troubleshooting agent, troubleshooting of a plurality of issues in a network indicated by a plurality of support tickets opened by users of the network. The device aggregates, based on results of the troubleshooting, the plurality of support tickets into a master support ticket. The device obtains a resolution to the master support ticket from a support engineer. The device uses the resolution to the master support ticket in conjunction with the large language model-based troubleshooting agent to troubleshoot a new support ticket.

Abstract: In one implementation, a device receives feedback from a user regarding an output of a large language model-based troubleshooting agent for a network that the large language model-based troubleshooting agent generates in response to an input request from the user. The device determines a network scenario associated with the input request and the output. The device causes, based on the feedback, the network scenario to be replicated in a test network. The device updates the large language model-based troubleshooting agent in part by using the large language model-based troubleshooting agent to assess the network scenario in the test network.

Abstract: In one implementation, a device makes a determination as to whether an issue identified in a network is eligible for self-healing. The device uses, based on the determination, a large language model-based agent to determine a root cause of the issue. The device identifies, using the large language model-based agent, a set of one or more self-healing actions to address the root cause of the issue. The device causes, using the large language model-based agent, performance of the set of one or more self-healing actions in the network.

Abstract: In one implementation, a device may determine how well a large language model-based troubleshooting agent for a network was able to perform during a first test having a first difficulty. The device may update the large language model-based troubleshooting agent using reinforcement learning based on how well the large language model-based troubleshooting agent was able to perform during the first test. The device may select a second difficulty for a second test based on how well the large language model-based troubleshooting agent was able to perform during the first test. The device may initiate the second test to assess how well the large language model-based troubleshooting agent is able to perform.

Abstract: In one implementation, a device receives a prompt for input to a large language model of a troubleshooting agent for a computer network. The device generates, using the troubleshooting agent, an intermediate answer by inputting the prompt to the large language model. The device identifies a hyperlink to a view associated with the intermediate answer and provided by an online application associated with administration of the computer network. The device provides the hyperlink in conjunction with the intermediate answer for display.

Abstract: In one implementation, a device uses a large language model-based agent to identify a task to correct an issue in a network. The device makes a determination that the large language model-based agent cannot complete the task. The device identifies, based on the determination, a subject matter expert to help complete the task. The device sends a request to the subject matter expert to complete the task.

Abstract: In one implementation, a device configures an impairment scenario in a computer network. The device performs an evaluation of the impairment scenario by a language model-based troubleshooting agent. The device makes a comparison between a root cause associated with the impairment scenario and the evaluation of the impairment scenario by the language model-based troubleshooting agent. The device provides an indication of the comparison.

Abstract: In one implementation, a device obtains a natural language-based description of a network via a user interface. The device generates, based on the natural language-based description, network configuration parameters for the network using a generative model. The device conducts a simulation of traffic in the network using the network configuration parameters, to obtain telemetry data. The device uses the telemetry data to train a machine learning model to perform network analytics.

Abstract: In one embodiment, a device causes, in accordance with a probing strategy, performance of a probing test by one or more agents in a network and with respect to an online application. The device obtains quality of experience measurements for the online application. The device adjusts, using reinforcement learning, the probing strategy based on how well a predictive model was able to predict the quality of experience measurements given results of the probing test. The device repeats the causing, obtaining, and adjusting steps using the probing strategy adjusted by the device, to find a minimally disruptive probing strategy that provides acceptable performance by the predictive model.

Abstract: In one embodiment, a device obtains data indicative of quality of experience for an online application. The device predicts, based on the data, path performances of network paths between an endpoint and the online application for different traffic loads. The device selects traffic loads for the network paths between the endpoint and the online application, based on the path performances predicted by the device. The device causes application traffic to be load balanced across the network paths between the endpoint and the online application, in accordance with those traffic loads selected by the device.

Abstract: In one implementation, a device generates a response to user input using a first language model. The device determines that the response is an erroneous response. The device generates a resolution to the erroneous response using a teacher language model. The device updates the first language model using the resolution from the teacher language model.

Abstract: In one implementation, a device uses a large language model associated with a network controller for a computer network to generate answers to questions regarding the computer network. The device determines that a particular answer to one of the questions represents a failure of the large language model. The device classifies the failure of the large language model as belonging to particular type of failure. The device provides an indication of the particular type of failure for display.

Abstract: In one implementation, a device pauses generation of computer code by a language model. The device matches a block of the computer code to a hint regarding a portion of the block of computer code. The device inserts the hint into the computer code. The device resumes generation of the computer code by the language model, wherein the language model uses the hint to generate a new portion of the computer code.

Abstract: In one implementation, a device obtains a prompt from a source for input to a language model. The device performs a search for similar prompts to that of the prompt in a prompt database. The device determines, based on results of the search, whether the prompt represents a potential attack. The device prevents a response to the prompt from being returned from the language model to the source, when the device determines that the prompt represents a potential attack.

Abstract: In one embodiment, a supervisory service for a network obtains quality of experience metrics for application sessions of an online application. The supervisory service maps the application sessions to paths that traverse a plurality of autonomous systems. The supervisory service identifies, based in part on the quality of experience metrics, a particular autonomous system from the plurality of autonomous systems associated with a decreased quality of experience for the online application. The supervisory service causes application traffic for the online application to avoid the particular autonomous system.

Abstract: In one embodiment, a service tracks performance of a machine learning model over time. The machine learning model is used to monitor one or more computer networks based on data collected from the one or more computer networks. The service also tracks performance metrics associated with training of the machine learning model. The service determines that a degradation of the performance of the machine learning model is anomalous, based on the tracked performance of the machine learning model and performance metrics associated with training of the model. The service initiates a corrective measure for the degradation of the performance, in response to determining that the degradation of the performance is anomalous.

Abstract: In one embodiment, a device identifies, based on traceroute information for a path in a network between an endpoint client and an online application, a particular segment of the path as most likely to cause degraded performance along the path. The device makes, using a prediction model, a prediction that routing traffic for the online application via the path will result in degraded quality of experience for the online application. The device obtains, based on the prediction, additional traceroute information in the network, to identify a bypass path in the network between the endpoint client and the online application that bypasses the particular segment. The device causes traffic for the online application to be routed along the bypass path.