Agentic Software Engineering: Foundational Pillars and a Research Roadmap
Ahmed E. Hassan, Hao Li, Dayi Lin, Bram Adams, Tse-Hsun Chen, Yutaro Kashiwa, Dong Qiu
2025-09-19
Summary
This paper introduces a new way of thinking about software engineering, called Agentic Software Engineering, where AI agents aren't just writing code, but actually working towards bigger software goals. It highlights the need to consider how humans and agents will work *together* in this new world, and proposes tools to make that collaboration happen.
What's the problem?
Currently, software engineering relies heavily on human developers. As we start using AI agents to help with more complex tasks, we don't have a clear framework for how humans and agents should interact to build trustworthy software. Simply having agents generate code isn't enough; we need a system where agents can handle tasks independently but also know when to ask for help from humans, and vice-versa. The existing methods for software development don't really account for this new dynamic.
What's the solution?
The paper proposes a 'Structured Agentic Software Engineering' (SASE) approach built around two main environments. The 'Agent Command Environment' (ACE) is where humans oversee and guide teams of AI agents, receiving packaged results and requests for advice. The 'Agent Execution Environment' (AEE) is where the agents actually do the work, and can automatically ask humans for help when they encounter difficult problems. This creates a back-and-forth system, redefining how humans and AI collaborate on software projects, moving beyond just 'agentic coding' to full 'agentic software engineering'.
Why it matters?
This is important because AI is rapidly changing the field of software engineering. If we don't develop a structured way for humans and AI to work together, we risk creating unreliable or untrustworthy software. This paper provides a starting point for a community-wide discussion about how to build this future, and also considers how software engineering education needs to adapt to prepare developers for this new era.
Abstract
Agentic Software Engineering (SE 3.0) represents a new era where intelligent agents are tasked not with simple code generation, but with achieving complex, goal-oriented SE objectives. To harness these new capabilities while ensuring trustworthiness, we must recognize a fundamental duality within the SE field in the Agentic SE era, comprising two symbiotic modalities: SE for Humans and SE for Agents. This duality demands a radical reimagining of the foundational pillars of SE (actors, processes, tools, and artifacts) which manifest differently across each modality. We propose two purpose-built workbenches to support this vision. The Agent Command Environment (ACE) serves as a command center where humans orchestrate and mentor agent teams, handling outputs such as Merge-Readiness Packs (MRPs) and Consultation Request Packs (CRPs). The Agent Execution Environment (AEE) is a digital workspace where agents perform tasks while invoking human expertise when facing ambiguity or complex trade-offs. This bi-directional partnership, which supports agent-initiated human callbacks and handovers, gives rise to new, structured engineering activities (i.e., processes) that redefine human-AI collaboration, elevating the practice from agentic coding to true agentic software engineering. This paper presents the Structured Agentic Software Engineering (SASE) vision, outlining several of the foundational pillars for the future of SE. The paper culminates in a research roadmap that identifies a few key challenges and opportunities while briefly discussing the resulting impact of this future on SE education. Our goal is not to offer a definitive solution, but to provide a conceptual scaffold with structured vocabulary to catalyze a community-wide dialogue, pushing the SE community to think beyond its classic, human-centric tenets toward a disciplined, scalable, and trustworthy agentic future.