ALMAS: an Autonomous LLM-based Multi-Agent Software Engineering Framework

2025-10-03

AI Summary

Existing LLM-based multi-agent systems predominantly focus on code generation, failing to comprehensively support the full software development lifecycle (SDLC). Method: This paper introduces AgileAgent—a self-organizing multi-agent software engineering framework designed for agile development. It models SDLC phases through role-aligned agents (e.g., Product Owner, Developer, Tester), enabling end-to-end collaboration across requirements analysis, coding, testing, and maintenance. The framework employs a role-driven architecture, modular integration interfaces, and lightweight agile process modeling, ensuring seamless interoperability with human developers and existing tools (e.g., IDEs). Contribution/Results: Experiments demonstrate that AgileAgent autonomously generates runnable applications and incrementally implements new features via iterative refinement. It significantly improves human-AI collaboration efficiency and task completion completeness. AgileAgent establishes a scalable, role-aware paradigm for LLM-augmented full-stack software engineering, bridging critical gaps between AI capabilities and industrial SDLC practices.

Technology Category

• Multiagent Systems: Teamwork
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

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.

Problem

• Automating multiple software development lifecycle stages using LLM agents
• Integrating autonomous agents with human developers in agile teams
• Creating modular multi-agent framework for end-to-end software engineering tasks

Innovation

• Autonomous multi-agent framework for software engineering
• Aligns LLM agents with agile development roles
• Modular integration with human developers and environments