CardioDev-Guard

CardioDev-Guard is an AI-assisted project quality and risk analysis platform designed to help developers identify potential problems in a project before release. Instead of relying only on manual code review and scattered testing, the platform brings multiple quality checks into a unified analysis workflow. The system analyzes a project across several dimensions, including missing values, duplicate data, class imbalance, data leakage, model metrics, and overall quality indicators. Its modular architecture combines specialized analyzers with audit adapters, a central scanning pipeline, findings aggregation, explanations, recommendations, and reporting. The target users are students, developers, ML engineers, and project teams who need a practical way to validate an AI/ML project before deployment or submission. A user can provide a project and run the analysis through the application, review detected findings, understand why an issue matters, and use the generated recommendations to improve the project. A key aspect of CardioDev-Guard is its focus on turning raw analysis results into actionable developer feedback rather than simply displaying test outputs. The system organizes findings by type and severity and supports repeatable quality checks through dedicated analyzers and tests. IBM Bob was used throughout the development workflow as an AI development partner for project ideation, implementation, debugging, testing, code analysis, documentation, and refinement. The team used Bob to work across the codebase, create and improve components, investigate issues, generate tests, and iterate on the final implementation. The project demonstrates how AI-assisted development can be combined with structured software-quality checks to make project validation more systematic, explainable, and easier to review before release.