IP Library › Granted Patent US 12,632,236
Granted Patent B2
US 12,632,236 · App. 18/440,413 · Granted May 19, 2026

Custom code assistant system facilitating code conversion methodology

Inventors: Krishnaveni Kavuri (Waxhaw, NC); Daniel Caricato (Fort Mill, SC); Bjorn Austraat (New York, NY)
Assignee: TRUIST BANK
G06F8/447G06F8/51G06F8/60G06F11/3688G06F21/31G06F21/629G06F40/30
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,632,236
App. No.
18/440,413
Filed
Feb 13, 2024
Granted
May 19, 2026
Kind
B2
Art Unit
2407
USPC
726/1
Abstract

Systems and methods obtain a translation request accompanying source code of source file(s) for conversion. Formatting of the source code of the source file(s) is verified and the source code is converted to target code that includes target file(s). The converting (i) maps programming language functions of the source code to corresponding programming language functions of the target code and (ii) utilizes, through customized prompt template(s), one or more large language models (LLMs) hosted by a cloud platform during the code conversion, where a source application that uses the source code includes TIBCO™ software. The target file(s) are distributed.

Claims (35)

1 . A computing system facilitating code conversion, comprising:

at least one processor;

a communication interface communicatively coupled to the at least one processor; and

a memory device storing executable code that, when executed, causes the at least one processor to:

obtain, via a web-based application from one or more user devices, a translation request accompanying source code of one or more source files for conversion;

verify formatting of the source code of the one or more source files;

convert the source code of the one or more source files to target code comprising one or more target files, the converting including: (i) mapping programming language functions of the source code to corresponding programming language functions of the target code, and (ii) utilizing, through one or more customized prompt templates, one or more large language models (LLMs) hosted by a cloud platform during the code conversion, wherein a source application that uses the source code comprises TIBCO™ software; and

distribute, via the web-based application, the one or more target files to the one or more user devices.

2 . The code assistant computing system of claim 1 , wherein the converting is initiated using a connector prior to converting flow components of the source code.

3 . The code assistant computing system of claim 1 , wherein the customized prompt templates initiate error handling management by causing the one or more LLMs to analyze how errors are handled in a source application using the source code and implementing similar error handling in a target application using the target code.

4 . The code assistant computing system of claim 1 , wherein the customized prompt templates initiate security configurations by causing the one or more LLMs to analyze how security is addressed in a source application using the source code and implementing similar security configurations in a target application using the target code.

5 . The code assistant computing system of claim 1 , wherein the programming language functions of the source code comprise at least one selected from the group consisting of a service class, an implementation type, operations, endpoint bindings, and a web services description language (WSDL).

6 . The code assistant computing system of claim 1 , wherein the mapping includes mapping endpoint bindings and transport configurations.

7 . The code assistant computing system of claim 1 , wherein the mapping includes verifying simple object access protocol (SOAP) configurations are accurately reflected by the corresponding programming language functions of the target code.

8 . The code assistant computing system of claim 1 , wherein the code conversion includes establishing separate flows and sub-flows, including DataWeave language transformations, for each corresponding programming language function of the corresponding programming language functions of the target code.

9 . The code assistant computing system of claim 1 , wherein the one or more customized prompt templates instruct the one or more LLMs to handle logic flow of the programming language functions of the source code.

10 . The code assistant computing system of claim 1 , wherein the one or more target files comprise one or more extensible markup language (XML) files.

11 . A computer-implemented method, comprising:

obtaining, via a web-based application from one or more user devices, a translation request accompanying source code of one or more source files for conversion;

verifying formatting of the source code of the one or more source files;

converting the source code of the one or more source files to target code comprising one or more target files, the converting including: (i) mapping programming language functions of the source code to corresponding programming language functions of the target code, and (ii) utilizing, through one or more customized prompt templates, the one or more LLMs during the code conversion, wherein a source application that uses the source code comprises TIBCO™ software; and

distributing, via the web-based application, the one or more target files to the one or more user devices.

12 . The computer-implemented method of claim 11 , wherein the converting is initiated using a connector prior to converting flow components of the source code.

13 . The computer-implemented method of claim 11 , wherein the customized prompt templates initiate error handling management by causing the one or more LLMs to analyze how errors are handled in a source application using the source code and implementing similar error handling in a target application using the target code.

14 . The computer-implemented method of claim 11 , wherein the customized prompt templates initiate security configurations by causing the one or more LLMs to analyze how security is addressed in a source application using the source code and implementing similar security configurations in a target application using the target code.

15 . The computer-implemented method of claim 11 , wherein the programming language functions of the source code comprise at least one selected from the group consisting of a service class, an implementation type, operations, endpoint bindings, and a web services description language (WSDL).

16 . The computer-implemented method of claim 11 , wherein the mapping includes mapping endpoint bindings and transport configurations.

17 . A non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that, when executed by a processor, cause the processor to:

obtain, via a web-based application from one or more user devices, a translation request accompanying source code of one or more source files for conversion;

verify formatting of the source code of the one or more source files;

convert the source code of the one or more source files to target code comprising one or more target files, the converting including: (i) mapping programming language functions of the source code to corresponding programming language functions of the target code, and (ii) utilizing, through one or more customized prompt templates, one or more large language models (LLMs) hosted by a cloud platform during the code conversion, wherein a source application that uses the source code comprises TIBCO™ software; and

distribute, via the web-based application, the one or more target files to the one or more user devices.

18 . The computer-readable storage medium of claim 17 , wherein the converting is initiated using a connector prior to converting flow components of the source code.

19 . The computer-readable storage medium of claim 17 , wherein the customized prompt templates initiate error handling management by causing the one or more LLMs to analyze how errors are handled in a source application using the source code and implementing similar error handling in a target application using the target code.

20 . The computer-readable storage medium of claim 17 , wherein the customized prompt templates initiate security configurations by causing the one or more LLMs to analyze how security is addressed in a source application using the source code and implementing similar security configurations in a target application using the target code.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 13, 2024
From: KAVURI, KRISHNAVENI; CARICATO, DANIEL; AUSTRAAT, BJORN
To: TRUIST BANK
Reel/Frame 066582/0154 →
Continuity (2)
Provisional Application 63618022 · Jan 5, 2024
Related Publication 20250224944A1 · Jul 10, 2025
References Cited (12)
US 7954084B2 · Arcaro · 2011 [cited by applicant]
US 8051410B2 · Marfatia · 2011 [cited by examiner]
US 9372837B2 · Alli · 2016 [cited by examiner]
US 9766868B2 · Allen · 2017 [cited by applicant]
US 11669332B1 · Aiyer · 2023 [cited by applicant]
US 12019992B2 · Bahrami · 2024 [cited by applicant]
US 12346673B1 · Gonugondla · 2025 [cited by applicant]
US 12353854B2 · Sikand · 2025 [cited by applicant]
US 20100017785A1 · Volkmann · 2010 [cited by examiner]
US 20140019116A1 · Lundberg · 2014 [cited by applicant]
US 20220171699A1 · Velammal · 2022 [cited by examiner]
US 20250258659A1 · Kunz · 2025 [cited by applicant]