Granted Patent B2

US 12,732,434 · App. 18/739,134 · Granted Sep 8, 2026

Inventors: Borislav Glozman (Ramat Gan, IL); Roy Segal (Mishmar-David, IL); Avi Chapnick (Ramat Gan, IL)

Assignee: AMDOCS DEVELOPMENT LIMITED

H04L41/16G06F40/30G06F40/58H04L41/5003

Abstract

Claims (44)

1 . A non-transitory computer-readable media storing computer instructions which when executed by one or more processors of a device cause the device to:

receive a natural language input defining an intent for a service to be instantiated in a network, the natural language input describing properties of the service including inputs to the service, functionality of the service, outputs of the service, and visual interfaces of the service;

process the natural language input, using a large language model (LLM), to translate the intent into a communication service capable of being orchestrated in a network;

decompose, by the LLM, the communication service into network services and network resources for fulfillment of the intent;

produce, by the LLM, requirements for ensuring that fulfillment of the intent continues over time;

orchestrate the communication service in the network by orchestrating the network services and network resources in the network; and

monitor the requirements of the intent to ensure that fulfillment of the intent continues over time.

2 . The non-transitory computer-readable media of claim 1 , wherein the intent is translated to the communication service by the LLM:

translating the intent into technical features, and

determining a communication service that will fulfill the technical features as the communication service capable of being orchestrated in a network.

3 . The non-transitory computer-readable media of claim 2 , wherein the technical features include technical requirements, capabilities, constraints, and policies.

4 . The non-transitory computer-readable media of claim 1 , wherein the communication service is comprised of at least one existing service.

5 . The non-transitory computer-readable media of claim 1 , wherein the communication service is comprised of at least one new service.

6 . The non-transitory computer-readable media of claim 5 , wherein the new service is designed by the LLM.

7 . The non-transitory computer-readable media of claim 1 , wherein Chain of Thought is used with the LLM.

8 . The non-transitory computer-readable media of claim 1 , wherein Tree of Thought is used with the LLM.

9 . The non-transitory computer-readable media of claim 1 , wherein Retrieval-Augmented Generation (RAG) is used with the LLM.

10 . The non-transitory computer-readable media of claim 1 , wherein the LLM is trained on communication service definitions included in at least one of:

historical data, or

synthetically generated data.

11 . The non-transitory computer-readable media of claim 10 , wherein the communication service definitions are virtual service representations.

12 . The non-transitory computer-readable media of claim 1 , wherein the LLM uses a vector database that stores embeddings generated from descriptors of services currently defined in an orchestration system.

13 . The non-transitory computer-readable media of claim 1 , wherein the LLM is aware of existing services and resources in an orchestration system, including their properties, relationships, capabilities, requirements, and policies.

14 . The non-transitory computer-readable media of claim 1 , wherein orchestrating the communication service in the network includes initiating a life-cycle-management process for the communication service.

15 . The non-transitory computer-readable media of claim 14 , wherein the life-cycle-management process is performed to fulfill the intent.

16 . The non-transitory computer-readable media of claim 1 , wherein the network services and network resources are orchestrated in the network for providing the communication service.

17 . A method, comprising:

at a computer system:

receiving a natural language input defining an intent for a service to be instantiated in a network, the natural language input describing properties of the service including inputs to the service, functionality of the service, outputs of the service, and

visual interfaces of the service;

processing the natural language input, using a large language model (LLM), to translate the intent into a communication service capable of being orchestrated in a network;

decomposing, by the LLM, the communication service into network services and network resources for fulfillment of the intent;

producing, by the LLM, requirements for ensuring that fulfillment of the intent continues over time;

orchestrating the communication service in the network by orchestrating the network services and network resources in the network; and

monitoring the requirements of the intent to ensure that fulfillment of the intent continues over time.

18 . A system, comprising:

a non-transitory memory storing instructions; and

one or more processors in communication with the non-transitory memory that execute the instructions to:

receive a natural language input defining an intent for a service to be instantiated in a network, the natural language input describing properties of the service including inputs to the service, functionality of the service, outputs of the service, and visual interfaces of the service;

process the natural language input, using a large language model (LLM), to translate the intent into a communication service capable of being orchestrated in a network;

decompose, by the LLM, the communication service into network services and network resources for fulfillment of the intent;

produce, by the LLM, requirements for ensuring that fulfillment of the intent continues over time;

orchestrate the communication service in the network by orchestrating the network services and network resources in the network; and

monitor the requirements of the intent to ensure that fulfillment of the intent continues over time.

Assignments (1)

Continuity (1)

References Cited (14)