Amazon Kiro: Transforming Software Development Through AI-Powered Autonomous Coding

Ravi Kant Sharma

Abstract: The integration of AI-powered development assistants into the Software Development Lifecycle (SDLC) has shown promising results in improving developer productivity. This paper proposes a comprehensive research framework to evaluate Amazon Kiro, a next-generation autonomous AI coding assistant, through controlled experimentation with professional developers. We present a detailed methodology for measuring task completion time, code quality, cognitive load, and developer satisfaction across multiple SDLC phases including coding, testing, documentation, and code review. Based on preliminary observations and existing literature on AI coding assistants, we hypothesize significant productivity improvements while maintaining code quality standards. The proposed study design includes 15 software developers over a 4-week period using within-subjects experimental methodology. This framework provides a rigorous approach for empirically validating AI-assisted development tools and establishes a governance model for sustainable organizational adoption. The research design can be adapted for evaluating other autonomous AI development assistants.

Keywords: Amazon Kiro, AI Development, Software Lifecycle, Autonomous Coding, Developer Productivity

How to Cite?: Ravi Kant Sharma, "Amazon Kiro: Transforming Software Development Through AI-Powered Autonomous Coding", Volume 14 Issue 11, November 2025, International Journal of Science and Research (IJSR), Pages: 1150-1154, https://www.ijsr.net/getabstract.php?paperid=SR251115051642, DOI: https://dx.doi.org/10.21275/SR251115051642