# US 2026/0064410 A1 · Method and system for automating peer code reviews

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

- Number: US 2026/0064410 A1
- Document ID: US20260064410A1
- Status: Published application
- Kind: A1
- Application number: 18/938,981
- Filed: 2024-11-06
- Published: 2026-03-05
- Earliest claimed: 2024-09-05
- Assignees: JPMorgan Chase Bank, N.A.
- Applicants: JPMorgan Chase Bank, N.A.
- Inventors: Venkata Mohit TAMANAMPUDI; Ravi KURUGANTHY; Jayaprakash MOSES; Priya Darshini AMARA
- Web page: https://patentgrep.com/patent/US20260064410A1

## CPC

- G06F 8/73: Program documentation
- G06F 8/75: Structural analysis for program understanding
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method and a system for automating a peer code review are provided. The method includes: receiving a pull request associated with an evaluation of a source code; generating, based on the pull request, a workflow associated with the evaluation of the source code; transmitting the workflow and the source code to a plurality of AI agents; performing, via the plurality of AI agents, a review of the source code, and each respective AI agent of the plurality of AI agents is responsible for a separate code review process from among a plurality of code review processes; aggregating each respective result from among the plurality of code review processes; and determining, based on the aggregating of each respective result, whether the pull request passes the evaluation of the source code.

## Claims

### Claim 1

A method for automating a peer code review, the method being implemented by at least one processor, the method comprising:
receiving, by the at least one processor, a pull request associated with an evaluation of a source code;
generating, by the at least one processor and based on the pull request, a workflow associated with the evaluation of the source code;
transmitting, by the at least one processor, the workflow and the source code to a plurality of AI agents;
performing, by the at least one processor via the plurality of AI agents, a review of the source code, wherein each respective AI agent of the plurality of AI agents is responsible for a separate code review process from among a plurality of code review processes;
aggregating, by the at least one processor, each respective result from among the plurality of code review processes; and
determining, by the at least one processor and based on the aggregating of each respective result, whether the pull request passes the evaluation of the source code.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
coordinating, by the at least one processor via a Large Language Model (LLM), each of the plurality of the AI agents and providing context-aware processing for the performing of the review, wherein the LLM is trained on at least one from among a code framework and an engineer handbook in order to understand and generate responses in a context associated with a software development.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
identifying, by the at least one processor via the LLM, at least one potential problem associated with the source code; and
generating, by the at least one processor via the LLM, at least one proposed source code solution, based on the at least one from among the code framework and the engineer handbook, for remedying the at least one identified potential problem.

### Claim 4 (depends on 2)

The method of claim 2, further comprising:
generating, by the at least one processor via the LLM, at least one from among a first explanation that relates to an understanding of the code framework and a second explanation that relates to at least one engineering concept associated with the source code.

### Claim 5 (depends on 1)

The method of claim 1, wherein the plurality of code review processes includes at least one from among validating a pipeline configuration, testing an individual module of the source code, testing a contract of the source code, testing a component of the source code, testing acceptance of the source code, testing end-to-end results of the source code, testing a performance of the source code, testing resiliency of the source code, providing code review feedback of the source code, testing a compatibility of the source code, testing security vulnerabilities of the source code, and testing business functionality of the source code.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
triggering, by the at least one processor via a developer platform, the generating of the workflow; and
coordinating, by the at least one processor via the developer platform, the performing of the review by the plurality of AI agents.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
integrating, by the at least one processor, at least one source code analysis tool for testing the source code and for validating the evaluation of the source code.

### Claim 8 (depends on 1)

The method of claim 1, wherein each of the plurality of code review processes is executed in parallel among the plurality of AI agents.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
generating, by the at least one processor, a report that includes the determination of whether the pull request passes the evaluation, an assessment of each respective result from among the plurality of code review processes, and an assessment of compliance of the source code with a predetermined standard; and
transmitting, by the at least one processor, the report to a user associated with the pull request.

### Claim 10

A computing apparatus for automating peer code reviews, the computing apparatus comprising:
a processor;
a memory; and
a communication interface coupled to each of the processor and the memory,
wherein the processor is configured to:
receive a pull request associated with an evaluation of a source code;
generate, based on the pull request, a workflow associated with the evaluation of the source code;
transmit the workflow and the source code to a plurality of AI agents;
perform, via the plurality of AI agents, a review of the source code, wherein each respective AI agent of the plurality of AI agents is responsible for a separate code review process from among a plurality code review processes;
aggregate each respective result from among the plurality of code review processes; and
determine, based on the aggregating of each respective result, whether the pull request passes the evaluation of the source code.

### Claim 11 (depends on 10)

The computing apparatus of claim 10, wherein the processor is further configured to:
coordinate, via a Large Language Model (LLM), each of the plurality of the AI agents and providing context aware processing for the performing of the review, wherein the LLM is trained on at least one from among a code framework and an engineer handbook in order to understand and generate responses in a context associated with a software development.

### Claim 12 (depends on 11)

The computing apparatus of claim 11, wherein the processor is further configured to:
identify, via the LLM, at least one potential problem associated with the source code; and
generate, via the LLM, at least one proposed source code solution, based on the at least one from among the code framework and the engineer handbook, for remedying the at least one identified potential problem.

### Claim 13 (depends on 11)

The computing apparatus of claim 11, wherein the processor is further configured to:
generate, via the LLM, at least one from among a first explanation that relates to an understanding of the code framework and a second explanation that relates to at least one engineering concept associated with the source code.

### Claim 14 (depends on 10)

The computing apparatus of claim 10, wherein the plurality of code review processes includes at least one from among validating a pipeline configuration, testing an individual module of the source code, testing a contract of the source code, testing a component of the source code, testing acceptance of the source code, testing end-to-end results of the source code, testing a performance of the source code, testing resiliency of the source code, providing code review feedback of the source code, testing a compatibility of the source code, testing security vulnerabilities of the source code, and testing business functionality of the source code.

### Claim 15 (depends on 10)

The computing apparatus of claim 10, wherein the processor is further configured to:
trigger, via a developer platform, the generation of the workflow; and
coordinate, via the developer platform, the performance of the review by the plurality of AI agents.

### Claim 16 (depends on 10)

The computing apparatus of claim 10, wherein the processor is further configured to:
integrate at least one source code analysis tool for testing the source code and to validate the evaluation of the source code.

### Claim 17 (depends on 10)

The computing apparatus of claim 10, wherein each of the plurality of code review processes is executed in parallel among the plurality of AI agents.

### Claim 18 (depends on 10)

The computing apparatus of claim 10, wherein the processor is further configured to:
generate a report that includes the determination of whether the pull request passes the evaluation, an assessment of each respective result from among the plurality of code review processes, and an assessment of compliance of the source code with a predetermined standard; and
transmit the report to a user associated with the pull request.

### Claim 19

A non-transitory computer readable storage medium storing instructions for automating peer code reviews, the storage medium comprising executable code which, when executed by a processor, causes the processor to:
receive a pull request associated with an evaluation of a source code;
generate, based on the pull request, a workflow associated with the evaluation of the source code;
transmit the workflow and the source code to a plurality of AI agents;
perform, via the plurality of AI agents, a review of the source code, wherein each respective AI agent of the plurality of AI agents is responsible for a separate code review process from among a plurality code review processes;
aggregate each respective result from among the plurality of code review processes; and
determine, based on the aggregating of each respective result, whether the pull request passes the evaluation of the source code.

### Claim 20 (depends on 19)

The storage medium of claim 19, wherein when executed by the processor, the executable code further causes the processor to:
coordinate, via a Large Language Model (LLM), each of the plurality of the AI agents and providing context aware processing for the performing of the review, wherein the LLM is trained on at least one from among a code framework and an engineer handbook in order to understand and generate responses in a context associated with a software development.
