LLM-based agent as a back-office virtual network troubleshooting assistant
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.
1 . A method comprising:
performing, by a device and 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, wherein the device structures troubleshooting requests to a large language model in accordance with a policy that constrains resources allocated to a large language model-based troubleshooting agent's troubleshooting requests to the large language model;
aggregating, by the device and based on results of the troubleshooting, the plurality of support tickets into a master support ticket;
obtaining, by the device, a resolution to the master support ticket from a support engineer; and
using, by the device, the resolution to the master support ticket in conjunction with the large language model-based troubleshooting agent to troubleshoot a new support ticket.
2 . The method as in claim 1 , wherein the large language model-based troubleshooting agent is configured to perform corrective actions in the network for an issue indicated by a particular support ticket.
3 . The method as in claim 1 , wherein the results of the troubleshooting indicate an action to be performed by the users, and wherein the method further comprises:
sending a notification to the users to perform the action.
4 . The method as in claim 1 , wherein the device aggregates the plurality of support tickets into the master support ticket by applying clustering to the plurality of support tickets.
5 . The method as in claim 1 , wherein the results of the troubleshooting indicate a common element or elements in the network associated with the plurality of issues.
6 . The method as in claim 1 , wherein the device performs the troubleshooting of the plurality of issues using the large language model-based troubleshooting agent, in accordance with a policy that controls which types of issues the large language model-based troubleshooting agent is allowed to troubleshoot.
7 . The method as in claim 1 , wherein the policy controls a maximum number of tokens that the large language model-based troubleshooting agent is allowed to pass to the large language model to troubleshoot a particular issue.
8 . The method as in claim 1 , wherein the results of the troubleshooting indicate that the large language model-based troubleshooting agent was unable to identify a root cause of the plurality of issues.
9 . The method as in claim 1 , wherein the large language model-based troubleshooting agent generates the results of the troubleshooting based in part on a database of previously resolved support tickets.
10 . The method as in claim 1 , wherein the device aggregates the plurality of support tickets based on a configured maximum number of issues per master support ticket.
11 . An apparatus, comprising:
one or more network interfaces;
a processor coupled to the one or more network interfaces and configured to execute one or more processes; and
a memory configured to store a process that is executable by the processor, the process when executed configured to:
perform, 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, wherein the device structures troubleshooting requests to a large language model in accordance with a policy that constrains resources allocated to a large language model-based troubleshooting agent's troubleshooting requests to the large language model;
aggregate, based on results of the troubleshooting, the plurality of support tickets into a master support ticket;
obtain a resolution to the master support ticket from a support engineer; and
use the resolution to the master support ticket in conjunction with the large language model-based troubleshooting agent to troubleshoot a new support ticket.
12 . The apparatus as in claim 11 , wherein the large language model-based troubleshooting agent is configured to perform corrective actions in the network for an issue indicated by a particular support ticket.
13 . The apparatus as in claim 11 , wherein the results of the troubleshooting indicate an action to be performed by the users, and wherein the process when executed is further configured to:
send a notification to the users to perform the action.
14 . The apparatus as in claim 11 , wherein the apparatus aggregates the plurality of support tickets into the master support ticket by applying clustering to the plurality of support tickets.
15 . The apparatus as in claim 11 , wherein the results of the troubleshooting indicate a common element or elements in the network associated with the plurality of issues.
16 . The apparatus as in claim 11 , wherein the apparatus performs the troubleshooting of the plurality of issues using the large language model-based troubleshooting agent, in accordance with a policy that controls which types of issues the large language model-based troubleshooting agent is allowed to troubleshoot.
17 . The apparatus as in claim 11 , wherein the policy controls a maximum number of tokens that the large language model-based troubleshooting agent is allowed to pass to the large language model to troubleshoot a particular issue.
18 . The apparatus as in claim 11 , wherein the results of the troubleshooting indicate that the large language model-based troubleshooting agent was unable to identify a root cause of the plurality of issues.
19 . The apparatus as in claim 11 , wherein the large language model-based troubleshooting agent generates the results of the troubleshooting based in part on a database of previously resolved support tickets.
20 . A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
performing, by the device and 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, wherein the device structures troubleshooting requests to a large language model in accordance with a policy that constrains resources allocated to a large language model-based troubleshooting agent's troubleshooting requests to the large language model;
aggregating, by the device and based on results of the troubleshooting, the plurality of support tickets into a master support ticket;
obtaining, by the device, a resolution to the master support ticket from a support engineer; and
using, by the device, the resolution to the master support ticket in conjunction with the large language model-based troubleshooting agent to troubleshoot a new support ticket.