Ordering infrastructure using application terms

A method, comprising:

• training, by a system comprising at least one processor, a generator neural network to produce a trained generator neural network, wherein the trained generator neural network is configured to output a bill of materials in response to receiving functional requirements for a computer system;
• training, by the system, a discriminator neural network to produce a trained discriminator neural network, wherein the trained discriminator neural network is configured to output whether the bill of materials received from the trained generator neural network satisfies the functional requirements for the computer system;
• and
• producing a resulting bill of materials from the trained generator neural network and the trained discriminator neural network based on a first identifier of functional requirements, wherein the producing comprises:
• the trained generator neural network of the generative adversarial network producing the output bill of materials, and
• after the trained generator neural network of the generative adversarial network produces the output bill of materials, the trained discriminator neural network of the generative adversarial network validating the output bill of materials relative to the functional requirements.

A non-transitory computer-readable medium comprising instructions that, in response to execution, cause a system comprising at least one processor to perform operations, comprising:

• training a first neural network to produce a trained first neural network, wherein the trained first neural network is configured to output a bill of materials in response to receiving functional requirements for a computer system;
• training a second neural network to produce a trained second neural network, wherein the trained second neural network is configured to output whether the bill of materials received from the trained first neural network satisfies the functional requirements for the computer system;
• and
• storing a generated bill of materials, wherein the trained first neural network produces the generated bill of materials based on the functional requirements, and wherein, after the trained first neural network produces the generated bill of materials, the trained second neural network validates the generated bill of materials against the functional requirements.