[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84788-en":3,"doc-seo-84788-105":30,"detail-sidebar-cat-0-en-105":95},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},84788,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","Three-Phase Evaluation of AI-Assisted Software Development Life Cycle","Three-Phase Evaluation of AI-Assisted Software Development Life Cycle examines how increasing AI autonomy changes software development outcomes, focusing on productivity, requirement adherence, and developer cognitive workload. Four developers reimplemented the same full-stack web application in three phases: partial AI assistance with GitHub Copilot, an AI-exclusive Copilot workflow, and an AI-exclusive workflow with AWS Kiro. Metrics include development effort, RITM requirement adherence, AI interaction efficiency, and NASA-TLX workload. Higher autonomy reduced effort and mental workload while modestly increasing frustration; AWS Kiro showed the best overall performance across most dimensions.","Three-Phase Evaluation of AI-Assisted Software Development Life Cycle  \nDr. Joshua Strübel, Professor Carrie Russell, Carson Crockett, Jason Ferraro, Nathan Londhe, Uzayr Syed, and Jacob Viehe  \nAbstract-This paper presents an exploratory evaluation of how increasing levels of AI autonomy affect software development productivity, requirement adherence, and developer cognitive workload. A team of four developers reimplemented the same full-stack web application across three sequential phases: partial AI-assisted development using GitHub Copilot, an AI-exclusive workflow using GitHub Copilot, and an AI-exclusive workflow using AWS  \nKiro. Evaluation metrics included development effort (hours), requirement adherence (RITM score), AIinteraction efficiency, and NASA-TLX workload measures. Across phases, higher levels of AI autonomy were associated with reduced development effort, improved requirement adherence, and lower self-reported mental workload, while developer frustration increased modestly. The AWS Kiro phase achieved the strongest overall performance on most measured dimensions, suggesting that tooling architecture may influence outcomes independently of AI autonomy level. Index Terms—AI-assisted software engineering, agentic AI, software productivity, developer cognition, human-AI collaboration, GitHub Copilot, AWSKiro  \nI. INTRODUCTION  \nThe proliferation of AI-powered coding assistants has substantially changed how software engineers approach software development. Tools such as GitHub Copilot and Amazon Kiro have evolved from supplemental systems into autonomous agents capable of planning, implementing, and iterating on entire features with minimal human intervention [7] . GitHub Copilot operates as an in-editor assistant that generates code suggestions, and in its agentic mode, can autonomously implement multi-step features from natural language prompts, and detect and resolve bugs without specific instruction. Amazon Kiro, a spec-driven development tool, takes amore structured approach: developers author formal specification documents, and the agent decomposes and implements features with minimal iterative prompting. Together, these platforms represent meaningfully different architectural philosophies for human-AI collaboration in software development. As adoption accelerates, organizations face a practical question:  \nbeyond simple code completion, how do different levels of AI autonomy and agentic platforms affect measurable development outcomes?  \nPrior controlled research has established that AIassisted development can yield meaningful productivity gains. Peng et al. found that developers using GitHub Copilot completed an [HTTP server](HTTP server) implementation 55.8% faster than a control group [1], and larger field experiments have corroborated these results across diverse  \nengineering contexts [2]. However, most existing work focuses on partial AI assistance rather than fully agentic scenarios in which developers serve primarily as orchestrators rather than implementers [7] . Published controlled empirical comparisons of competing agentic platforms under matched conditions remain limited at the time of writing.  \nThis study addresses these gaps through a three-phase controlled evaluation in which the same development team reimplemented the same application with systematically  \nvaried AI assistance levels and tooling. The primary contributions are:  \n• A controlled, multi-phase empirical comparison of partial and fully agentic AI-assisted development with respect to productivity, quality, and cognitive load dimensions.  \n• A quantitative comparison of two commercially  \navailable agentic platforms (GitHub Copilot and AWS Kiro) under equivalent task conditions.  \n• A replicable evaluation framework for future comparative studies of AI coding tools.  \nII. RELATED WORK  \nAI-Assisted Software Development  \nRecent research has demonstrated that AI-assisted programming tools can significantly improve developer prod","cbCaiqeVQfY1eTWU","https://ap.wps.com/l/cbCaiqeVQfY1eTWU","pdf",419344,2,1,7,"English","en",105,"# Introduction\n# Related Work\n# Research Questions","[{\"question\":\"What three development phases does the study compare?\",\"answer\":\"The study reimplements the same full-stack web application in three phases: partial AI-assisted development using GitHub Copilot, an AI-exclusive workflow using GitHub Copilot, and an AI-exclusive workflow using AWS Kiro.\"},{\"question\":\"Which metrics are used to evaluate performance across phases?\",\"answer\":\"Evaluation includes development effort (hours), requirement adherence (RITM score), AI interaction efficiency, and NASA-TLX measures of cognitive workload.\"},{\"question\":\"How does higher AI autonomy affect the measured outcomes?\",\"answer\":\"Higher autonomy is associated with reduced development effort, improved requirement adherence, and lower self-reported mental workload, while developer frustration increases modestly.\"},{\"question\":\"Why does the AWS Kiro phase perform best overall?\",\"answer\":\"The results suggest that tooling architecture can influence outcomes independently of AI autonomy level, with AWS Kiro achieving the strongest overall performance on most measured dimensions.\"}]",1784198242,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":90,"head_meta":92,"extra_data":94,"updated_unix":28},"three-phase-evaluation-of-ai-assisted-software-development-life-cycle","",{"@graph":36,"@context":89},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/three-phase-evaluation-of-ai-assisted-software-development-life-cycle/84788/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-22","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"What three development phases does the study compare?","Question",{"text":75,"@type":76},"The study reimplements the same full-stack web application in three phases: partial AI-assisted development using GitHub Copilot, an AI-exclusive workflow using GitHub Copilot, and an AI-exclusive workflow using AWS Kiro.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which metrics are used to evaluate performance across phases?",{"text":80,"@type":76},"Evaluation includes development effort (hours), requirement adherence (RITM score), AI interaction efficiency, and NASA-TLX measures of cognitive workload.",{"name":82,"@type":73,"acceptedAnswer":83},"How does higher AI autonomy affect the measured outcomes?",{"text":84,"@type":76},"Higher autonomy is associated with reduced development effort, improved requirement adherence, and lower self-reported mental workload, while developer frustration increases modestly.",{"name":86,"@type":73,"acceptedAnswer":87},"Why does the AWS Kiro phase perform best overall?",{"text":88,"@type":76},"The results suggest that tooling architecture can influence outcomes independently of AI autonomy level, with AWS Kiro achieving the strongest overall performance on most measured dimensions.","https://schema.org",{"og:url":51,"og:type":91,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":93,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":96},[97,101,105,109,114,119,123,126,131,134,138],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Exam",70,"exam",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},5,"Comic",60,"comic",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},6,"Technology",50,"technology",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":124,"slug":125},30,"research-report",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":129,"slug":130},9,"Religion & Spirituality",20,"religion-spirituality",{"id":129,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":129,"slug":133},"World Cup","world-cup",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":135,"slug":137},10,"Lifestyle","lifestyle",{"id":139,"doc_module":4,"doc_module_name":46,"category_name":140,"show_sort_weight":110,"slug":141},19,"General","general"]