Sprint Planning and Al/Ml: How to Balance Iterations with Data Complexity

Abstract

Purpose: This study aims to bridge the gap in modern software development by integrating agile methodologies with artificial intelligence and machine learning (AI/ML). It seeks to understand how agile sprint planning can effectively interact with the complexities of data inherent in AI/ML projects.

Methodology: The approach taken in this research draws upon a hermeneutic philosophy, which is inference-based, alongside a descriptive methodology. It investigates effective strategies for data preparation, model building, and validation, utilizing the iterative architecture of agile sprinting. The study also incorporates a critical examination to address significant limitations encountered in the integration process.

Findings: The findings highlight the essential role of cross-departmental teams and identify various technological tools that facilitate the smooth integration of agile methodologies with AI/ML projects. A rigorous examination also emphasizes the necessity for ongoing validation through evidence to manage the complexities effectively.

Unique Contribution to Theory, Practice, and Policy (Recommendations): The study offers a comprehensive framework and practical recommendations for businesses aiming to handle data-driven AI/ML initiatives in agile environments. It provides a strategic management approach that aligns more successfully with the demands of data production and agile processes, thus contributing significantly to both theoretical perspectives and practical applications in software development. These contributions are pivotal for informing policy on the integration of cutting-edge technologies in agile settings.