Home / Current Issue / Paper 1719198
Gen-AI Code Reviewer: An AI-Assisted Pull Request Review Platform for Faster First-Pass Software Quality Analysis
Subject area: Science,Engineering and Technology · Area of research: Generative AI
DOI: https://doi.org/10.64388/IREV9I12-1719198
Abstract
Code review remains one of the most important quality-assurance activities in modern software engineering, yet manual review of pull requests is often delayed by reviewer workload, inconsistent practices, and growing repository activity. This paper presents Gen-AI Code Reviewer, a full-stack web platform that automates first-pass pull request analysis using a large language model integrated with GitHub-based development workflows. The system authenticates users through GitHub OAuth, retrieves repository and pull request metadata through the GitHub REST API, extracts changed-file diffs, and submits structured review prompts to an AI model for contextual analysis. The generated output includes a summary, risk score, severity- tagged comments, and suggested fixes, which are persisted in a PostgreSQL database and displayed through a dashboard interface. The platform is implemented using Next.js, TypeScript, Prisma ORM, tRPC, React Query, Better Auth, and Inngest for asynchronous workflow execution. The proposed system is designed to reduce review turnaround time, improve consistency of early feedback, and support human reviewers by surfacing potentially risky changes before manual inspection. The paper describes the motivation, architecture, workflow, implementation strategy, and practical relevance of the prototype, and discusses limitations related to model reliability, repository context, and prompt-length constraints.
Keywords
Code Review, Pull Request Analysis, Generative AI, Github, Large Language Models, Software Quality, Developer Tooling, Next.Js
How to cite this paper
@article{1719198,
author = {Shahu Sakhare , Om Patil, Aditya Muke, Faizahmed Pathan , Vedang Panditrao },
title = {Gen-AI Code Reviewer: An AI-Assisted Pull Request Review Platform for Faster First-Pass Software Quality Analysis},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {12},
pages = {2705-2709},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1719198.pdf},
abstract = {Code review remains one of the most important quality-assurance activities in modern software engineering, yet manual review of pull requests is often delayed by reviewer workload, inconsistent practices, and growing repository activity. This paper presents Gen-AI Code Reviewer, a full-stack web platform that automates first-pass pull request analysis using a large language model integrated with GitHub-based development workflows. The system authenticates users through GitHub OAuth, retrieves repository and pull request metadata through the GitHub REST API, extracts changed-file diffs, and submits structured review prompts to an AI model for contextual analysis. The generated output includes a summary, risk score, severity- tagged comments, and suggested fixes, which are persisted in a PostgreSQL database and displayed through a dashboard interface. The platform is implemented using Next.js, TypeScript, Prisma ORM, tRPC, React Query, Better Auth, and Inngest for asynchronous workflow execution. The proposed system is designed to reduce review turnaround time, improve consistency of early feedback, and support human reviewers by surfacing potentially risky changes before manual inspection. The paper describes the motivation, architecture, workflow, implementation strategy, and practical relevance of the prototype, and discusses limitations related to model reliability, repository context, and prompt-length constraints.},
keywords = {Code Review, Pull Request Analysis, Generative AI, Github, Large Language Models, Software Quality, Developer Tooling, Next.Js},
month = {June},
doi = {https://doi.org/10.64388/IREV9I12-1719198}
}