# US 12,761,885 B2 · Generative artificial intelligence (“AI”) for development task feedback system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,761,885 B2 - Document ID: US12761885B2 - Status: Granted - Kind: B2 - Application number: 18/794,129 - Filed: 2024-08-05 - Published: 2026-02-05 - Granted: 2026-10-13 - Earliest claimed: 2024-08-05 - Assignees: Bank of America Corporation - Applicants: Bank of America Corporation - Inventors: Ramesh Bhavana; Rajneesh Acharya; Gaurav Mongia; Daniel E. Bartlett; Ramakrishna Yannam - Web page: https://patentgrep.com/patent/US12761885B2 ## CPC - G06F 8/35: model driven - G06F 8/33: Intelligent editors - G06F 8/73: Program documentation - G06N 3/0475: Generative networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Methods for harnessing GenAI to provide dynamic feedback to developers are provided. Methods may receive processed data elements. Each data element may include two or more iterations of a software element generated by a developer, and a feedback document generated by a tester in response to receiving the software element. Methods may train an LLM with the data elements. The LLM may operate with an AI feedback engine. Methods may receive a software element created by a developer. Methods may push the software element to the engine. Methods may assess the software element at the engine to generate the feedback document. The feedback document may include comments, modifications and/or a quality index. Methods may provide the feedback document to the developer. The developer may override the feedback document. Upon receipt of an override, the engine may send an unedited version of the software element to publication. ## Claims ### Claim 1 A method for harnessing generative artificial intelligence (“GenAI”) to provide dynamic feedback to developers, the method comprising: receiving a plurality of processed data elements, each processed data element, included in the plurality of the processed data elements, comprising two or more iterations of: a software element generated by a developer; and a feedback document generated by a tester in response to receiving the software element; training a large language model with the plurality of processed data elements, said large language model operating in tandem with an artificially intelligent feedback engine; enabling a first developer to create a first software element; pushing the first software element to the artificially intelligent feedback engine; assessing the first software element at the artificially intelligent feedback engine, said assessing comprising using the large language model operating in tandem with the artificially intelligent feedback engine to generate the feedback document for the first software element, said feedback document comprising: one or more comments on the first software element; one or more modifications for the first software element; and a quality index for the first software element; providing the feedback document to the first developer; enabling the first developer to: override the feedback document; and send an unedited version of the first software element to publication, said unedited version of the first software element absent incorporation of the one or more comments and/or one or more modifications. ### Claim 2 (depends on 1) The method of claim 1 wherein the first software element is a story, said story comprising a paragraph that identifies a functional or technical description of a system behavior. ### Claim 3 (depends on 1) The method of claim 1 wherein the first software element is a code segment. ### Claim 4 (depends on 1) The method of claim 1 wherein the first software element is a test case. ### Claim 5 (depends on 1) The method of claim 1, wherein the method further comprises: continually harvesting updates to correspondence between different entity levels regarding issues with publication of software elements; and continually updating the large language model with the updates. ### Claim 6 (depends on 1) The method of claim 1, wherein upon receipt of the override at the artificially intelligent feedback engine, the engine transmits an unedited version of the first software element to a production environment for publication. ### Claim 7 A system for harnessing generating artificial intelligence (“GenAI”) to provide dynamic feedback to developers, the system comprising: a hardware processor comprising an artificially intelligent feedback engine operating in tandem with a large language model, said large language model trained on: a plurality of processed data elements, each processed data element, included in the plurality of processed data elements comprising two or more iterations of: a software element generated by a developer; and a feedback document generated by a tester in response to receiving the software element; a software development user interface, said software development user interface enabling a developer to: create the software element; send the software element for review; and publish the software element; the artificially intelligent feedback engine operable to: intercept the software element sent for review; assess the software element using the large language model to generate the feedback document for the software element; provide the feedback document to the developer via the software development user interface, said feedback document comprising: one or more comments on the software element; one or more modifications for the software element; and a quality index for the software element; upon receipt of the feedback document at the software development user interface, enable the developer to: override the feedback document; and send an unedited version of the software element to publication, said unedited version of the software element absent incorporation of the one or more comments and/or one or more modifications. ### Claim 8 (depends on 7) The system of claim 7 wherein the software element is a story, said story comprising a paragraph that identifies a functional or technical description of a system behavior. ### Claim 9 (depends on 7) The system of claim 7 wherein the software element is a code segment. ### Claim 10 (depends on 7) The system of claim 7 wherein the software element is a test case. ### Claim 11 (depends on 7) The system of claim 7 wherein the artificially intelligent feedback engine is further operable to: continually harvest updates to correspondence between different entity levels regarding issues with publication of software elements; and continually update the large language model with the updates. ### Claim 12 (depends on 11) The system of claim 11, wherein upon receipt of the override at the artificially intelligent feedback engine, the engine transmits an unedited version of the software element to a production environment for publication. ### Claim 13 One or more non-transitory computer-readable media storing computer-executable instructions which, when executed by a processor on a computer system, perform a method for harnessing generating artificial intelligence (“GenAI”) to provide dynamic feedback to developers, the method comprising: receiving a plurality of processed data elements, each processed data element, included in the plurality of the processed data elements, comprising two or more iterations of: a software element generated by a developer; and a feedback document generated by a tester in response to receiving the software element; training a large language model with the plurality of processed data elements, said large language model operating in tandem with an artificially intelligent feedback engine; receiving the software element, said software element created by a developer; pushing the software element to the artificially intelligent feedback engine; assessing the software element at the artificially intelligent feedback engine, said assessing comprising using the large language model operating in tandem with the artificially intelligent feedback engine to generate the feedback document for the software element, said feedback document comprising: one or more comments on the software element; one or more modifications for the software element; and a quality index for the software element; providing the feedback document to the developer; enabling the developer to: override the feedback document; and send an unedited version of the software element to publication, said unedited version of the software element absent incorporation of the one or more comments and/or one or more modifications. ### Claim 14 (depends on 13) The computer readable medium of claim 13 wherein the software element is a story, said story comprising a paragraph that identifies a functional or technical description of a system behavior. ### Claim 15 (depends on 13) The computer readable medium of claim 13 wherein the software element is a code segment. ### Claim 16 (depends on 13) The computer readable medium of claim 13 wherein the software element is a test case. ### Claim 17 (depends on 13) The computer readable medium of claim 13, wherein the method further comprises: continually harvesting updates to correspondence between different entity levels regarding issues with publication of software elements; and continually updating the large language model with the updates. ### Claim 18 (depends on 13) The computer readable medium of claim 13, wherein upon receipt of the override at the artificially intelligent feedback engine, the engine transmits an unedited version of the software element to a production environment for publication.