# US 2026/0099328 A1 · Computer-implemented method and computer system for software modernization

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0099328 A1
- Document ID: US20260099328A1
- Status: Published application
- Kind: A1
- Application number: 18/908,147
- Filed: 2024-10-07
- Published: 2026-04-09
- Earliest claimed: 2024-10-07
- Assignees: Accenture Global Solutions Limited
- Applicants: Accenture Global Solutions Limited
- Inventors: Sherin VARGHESE; Arun Vidyadharan; Priyanka Yadav; Akshay Ravindra; Vijeth Srinivas Hegde; Koushik M Vijayaraghavan
- Web page: https://patentgrep.com/patent/US20260099328A1

## CPC

- G06F 8/76: Adapting program code to run in a different environment; Porting
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Methods, systems, and computer-readable storage media for converting legacy software into modern software. The legacy software and any supporting content for conversion is received. Content types and software categories are determined based on the legacy software and the supporting content. Further, actions to be taken for the software categories are identified and foundation models are selected based on the content types and the actions, to support the conversion. For each of the actions to convert the legacy software into the modern software, a sequence of steps is generated. Further, a check is performed to determine if any necessary software to execute the sequence of steps for each of the actions is missing. If any necessary software is missing, a replacement software is generated for the missing software. Further, the sequence of steps for each of the actions is executed to generate the modern software.

## Claims

### Claim 1

A computer-implemented method for converting legacy software into modern software, comprising:
receiving the legacy software and any supporting content for conversion;
first identifying content types within the legacy software and the supporting content;
first determining one or more software categories for the legacy software and the supporting content;
second identifying actions to be taken for the determined software categories;
first selecting a plurality of foundation models, based on the identified content types and the identified actions, to support the conversion;
first generating a sequence of steps for each of the actions to convert the legacy software into the modern software, the sequence of steps including foundation model steps and non-foundation model steps;
second determining if any necessary software to execute the steps is missing;
second generating, in response to a positive result of the second determining, replacement software for the missing software; and
executing the sequence of steps for each of the actions to generate the modern software.

### Claim 2 (depends on 1)

The method of claim 1 wherein the content types include any of software code, images, non-software text, and/or audio.

### Claim 3 (depends on 1)

The method of claim 1, wherein the software categories include any of legacy language code, Infrastructure as Code, Dev-Ops as code, user interface code, and/or images.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
third determining, based on a size of the legacy software, whether the legacy software requires chunking before executing;
second selecting, in response to a positive result of the third determining, at least one chunking methodology appropriate for the legacy software; and
separating at least a portion of the legacy software into chunks per the selected at least one chunking methodology;
wherein the executing the sequence of steps comprises executing the sequence of steps on the chunks.

### Claim 5 (depends on 1)

The method of claim 1, wherein the second selecting at least one chunking methodology comprises selecting:
logical chunking for object oriented programming;
token based chunking for non-object oriented programming;
file size based chunking for audio files;
duration length based chunking for video files; and
image-token based chunking for image files.

### Claim 6 (depends on 1)

The method of claim 1, wherein the generating the sequence of steps further comprises:
obtaining from a global template repository, an existing series of steps appropriate for the conversion; and/or
submitting the software categories and at least some of the received supporting content to a template generation model and receiving in response at least some of the steps for each of the actions from the template generation model including the foundation model steps and the non-foundation model steps.

### Claim 7 (depends on 6)

The method of claim 6, further comprising:
generating prompts and associated roles and parameters for the foundation model steps included in the sequence of steps generated for each of the actions; and
storing, for each of the actions, the sequence of steps including the foundation model steps, the non-foundation model steps, and the prompts and the associated roles and parameters for the foundation model steps, as a template in the global template repository.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
determining, based on the identified actions, corresponding quality metrics to measure performance of the foundation model steps; and
the first generating the sequence of steps comprises selecting steps that satisfy the corresponding metrics.

### Claim 9 (depends on 1)

The method of claim 1, wherein executing the sequence of steps for each of the actions to generate the modern software comprises:
scheduling an order of performing the actions;
mapping chunks corresponding to at least a portion of the legacy software for each of the actions;
mapping the sequence of steps for each of the actions;
assigning, from the selected plurality of foundation models, at least one foundation model for the foundation model steps in the sequence of steps; and
executing, for each of the actions, the sequence of steps on the respective chunks using the assigned at least one foundation model.

### Claim 10 (depends on 9)

The method of claim 9, wherein assigning, from the selected plurality of foundation models, the at least one foundation model for the sequence of steps comprises assigning the at least one foundation model based on one or more of:
a model reputation score of each of the selected plurality of foundation models;
features of each of the selected plurality of foundation models;
budget and time parameters associated with the sequence of steps; and
an action associated with the sequence of steps.

### Claim 11 (depends on 9)

The method of claim 9, further comprising:
deriving quality scores for the results of execution of the foundation model steps included in the sequence of steps by measuring performance of results of execution of the foundation model steps using a respectively determined quality metrics;
determining if the quality scores fall below a predetermined threshold score;
determining, in response to determining that the quality scores fall below the predetermined threshold score, to initiate a retry function, wherein the retry function includes tuning hyperparameters of at least one foundation model assigned for the foundation model steps, retrying executions of the foundation model steps using the tuned at least one foundation model, and deriving the quality scores for the retried executions; and
performing the retry function iteratively till the quality scores fall above the predetermined threshold score or a number of retry counts is greater than a predetermined retry count.

### Claim 12 (depends on 11)

The method of claim 11, further comprising:
determining, in response to determining that the quality scores fall below the predetermined threshold score and the number of retry counts is greater than the predetermined retry count, to initiate a rephrase function, wherein the rephrase function includes rephrasing prompts generated for the foundation model steps, retrying executions of the foundation model steps using rephrased prompts, and deriving the quality scores for the retried executions; and
performing the rephrase function iteratively till the quality scores fall above the predetermined threshold score or a number of rephrase counts is greater than a predetermined rephrase count.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
performing, in response to determining that the quality scores fall below the predetermined threshold score after performing the retry function and the rephrase function, a fine-tuning of the at least one foundation model assigned for the foundation model steps; and
initiating performing of the retry function following the rephrase function, till the quality scores fall above the predetermined threshold score.

### Claim 14 (depends on 1)

The method of claim 1, further comprising:
generating graphs for visualizing the results of execution of the sequence of steps for each of the actions; and
creating a collaborative platform for accessing the results of execution of the sequence of steps.

### Claim 15

A non-transitory computer readable media storing instructions programmed to cooperate with electronic computer hardware in combination with software to perform operations for converting legacy software into modern software, the operations comprising:
receiving the legacy software and any supporting content for conversion;
first identifying content types within the legacy software and the supporting content;
first determining one or more software categories for the legacy software and the supporting content;
second identifying actions to be taken for the determined software categories;
first selecting a plurality of foundation models, based on the identified content types and the identified actions, to support the conversion;
first generating a sequence of steps for each of the actions to convert the legacy software into the modern software, the sequence of steps including foundation model steps and non-foundation model steps;
second determining if any necessary software to execute the steps is missing;
second generating, in response to a positive result of the second determining, replacement software for the missing software; and
executing the sequence of steps for each of the actions to generate the modern software.

### Claim 16 (depends on 15)

The non-transitory computer readable media of claim 15, the operations further comprising:
third determining, based on a size of the legacy software, whether the legacy software requires chunking before executing;
second selecting, in response to a positive result of the third determining, at least one chunking methodology appropriate for the legacy software; and
separating at least a portion of the legacy software into chunks per the selected at least one chunking methodology;
wherein the executing the sequence of steps comprises executing the sequence of steps on the chunks.

### Claim 17 (depends on 15)

The non-transitory computer readable media of claim 15, wherein the generating a sequence of steps further comprises:
obtaining from a global template repository, an existing series of steps appropriate for the conversion; and/or
submitting the software categories and at least some of the received supporting content to a template generation model and receiving in response at least some of the steps for each of the actions from the template generation model including the foundation model steps and the non-foundation model steps.

### Claim 18

A system, comprising:
a processor;
a non-transitory computer readable memory storing instructions programmed to cooperate with the processor to perform operations for converting legacy software into modern software, the operations comprising:
receiving the legacy software and any supporting content for conversion;
first identifying content types within the legacy software and the supporting content;
first determining one or more software categories for the legacy software and the supporting content;
second identifying actions to be taken for the determined software categories;
first selecting a plurality of foundation models, based on the identified content types and the identified actions, to support the conversion;
first generating a sequence of steps for each of the actions to convert the legacy software into the modern software, the sequence of steps including foundation model steps and non-foundation model steps;
second determining if any necessary software to execute the steps is missing;
second generating, in response to a positive result of the second determining, replacement software for the missing software; and
executing the sequence of steps for each of the actions to generate the modern software.

### Claim 19 (depends on 18)

The system of claim 18, the operations further comprising:
third determining, based on a size of the legacy software, whether the legacy software requires chunking before executing;
second selecting, in response to a positive result of the third determining, at least one chunking methodology appropriate for the legacy software; and
separating at least a portion of the legacy software into chunks per the selected at least one chunking methodology;
wherein the executing the sequence of steps comprises executing the sequence of steps on the chunks.

### Claim 20 (depends on 19)

The system of claim 19, wherein the generating a sequence of steps further comprises:
obtaining from a global template repository, an existing series of steps appropriate for the conversion; and/or
submitting the software categories and at least some of the received supporting content to a template generation model and receiving in response at least some of the steps for each of the actions from the template generation model including the foundation model steps and the non-foundation model steps.
