Multi-agent Large Language Model (LLM) systems have been leading the way in applied LLM research across a number of fields. One notable area is software development, where researchers have advanced the automation of code implementation, code testing, code maintenance, inter alia, using LLM agents. However, software development is a multifaceted environment that extends beyond just code. As such, a successful LLM system must factor in multiple stages of the software development life-cycle (SDLC). In this paper, we propose a vision for ALMAS, an Autonomous LLM-based Multi-Agent Software Engineering framework, which follows the above SDLC philosophy such that it may work within an agile software development team to perform several tasks end-to-end. ALMAS aligns its agents with agile roles, and can be used in a modular fashion to seamlessly integrate with human developers and their development environment. We showcase the progress towards ALMAS through our published works and a use case demonstrating the framework, where ALMAS is able to seamlessly generate an application and add a new feature.
AgentForge: Execution-Grounded Multi-Agent LLM Framework for Autonomous Software Engineering
International Research Center for Complexity Sciences(复杂科学国际研究中心);Hangzhou International Innovation Institute(杭州国际创新研究院);Beihang University(北京航空航天大学);College of Science, Mathematics and Technology(科学、数学与技术学院);Wenzhou-Kean University(温州-凯恩大学);Fakulti Teknologi Maklumat dan Komunikasi(信息技术与通信学院);Universiti Teknikal Malaysia Melaka(马来西亚Melaka理工大学);Computer Systems Engineering Department(计算机系统工程系);Sukkur IBA University(苏库尔IBA大学);University of Electronic Science and Technology of China(电子科学与技术大学)
Towards autonomous normative multi-agent systems for Human-AI software engineering teams
Decision Systems Lab University of Wollongong(伍伦贡大学决策系统实验室);Monash University(莫纳什大学);Macquarie University(麦考瑞大学);University of British Columbia(不列颠哥伦比亚大学)
Cognitive Platform Engineering for Autonomous Cloud Operations
IEEE Senior Texas USA(IEEE高级工程师,得克萨斯州,美国);Albertsons Companies California USA(Albertsons公司,加利福尼亚州,美国);Cleveland State University Ohio USA(克利夫兰州立大学,俄亥俄州,美国);IEEE Senior Massachusetts USA(IEEE高级工程师,马萨诸塞州,美国);IEEE Senior California USA(IEEE高级工程师,加利福尼亚州,美国)
Engineering Reliable Autonomous Systems: Challenges and Solutions
University of Manchester(曼彻斯特大学);University of Nottingham(诺丁汉大学);University of Modena and Reggio Emilia(摩德纳大学);University of Aberdeen(阿伯丁大学);Utrecht University(乌得勒支大学);University of Milano-Bicocca(米兰比可卡大学);University of York(约克大学);University of Malta(马耳他大学);ENSTA | Institut Polytechnique de Paris(巴黎高等先进技术学院);Vanderbilt University(范德堡大学);ONERA(法国航空航天研究院);Politecnico di Milano(米兰理工大学);Kernkonzept GmbH(Kernkonzept有限公司);Lancaster University Leipzig(兰卡斯特大学莱比锡校区);Gran Sasso Science Institute(格兰萨索科学研究所);Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院);University of Oslo(奥斯陆大学);Southwest University(西南大学);Aarhus University(奥胡斯大学);Newcastle University(纽卡斯尔大学);KU Leuven(鲁汶大学);Maynooth University(梅努斯大学)
Software Engineering for Intelligent and Autonomous Systems: Report from the GI Dagstuhl Seminar 18343
软件系统越来越多地用于以不确定环境、不断变化的需求和意外故障为特征的应用领域;突然的系统故障会引起严重的安全性、舒适性或收入损失问题。在操作期间,这些系统可能需要处理若干不可预测的情况,包括系统性能的变化、系统工作量的突然变化和组件故障。这些情况可能导致偏离期望的系统行为,并且需要系统行为、参数或架构的动态适应。通过使用通常用软件实现的闭环控制,智能和自主软件系统可以通过识别异常情况、分析替代适应选项以及最终自适应到合适的新配置来动态地自适应,而无需任何或有限的人类参与。本报告总结了SEfIAS GI Dagstuhl研讨会期间开展的研究,该研讨会为SEAMS,ICAC/ICAC,SASO,Self-Aware Computing和AAMAS研究社区的早期职业研究人员和从业人员提供了加强互动和合作的论坛。
Rethinking Autonomy: Preventing Failures in AI-Driven Software Engineering
Department of CSE Chandigarh University Mohali, India(计算机科学与工程系 奇纳格里大学 莫哈利,印度);Department of CST Tsinghua University Beijing, China(计算机科学与技术系 清华大学 北京,中国)
Industry Best Practices in Robotics Software Engineering
Jet Propulsion Laboratory, California Institute of Technology(喷气推进实验室,加州理工学院);Robert Bosch Corporate Research(罗伯特·博世企业研究院);Waymo;XITASO GmbH(XITASO有限公司);Mälardalen University(梅拉达伦大学);Vrije Universiteit Amsterdam(阿姆斯特丹自由大学);University of Stuttgart(斯图加特大学)