Skip to content
Conference

Multi-Agent Enterprise AI for Autonomous Software Engineering and Operational Intelligence

Oct 2026 · 2026 IEEE Conference on Generative AI for Secure Systems (GAISS) · pp. 1-4 · 0 citations · 9 references

Abstract

Large Language Models (LLMs) now assist with coding, testing, documentation, debugging, and operations, yet enterprise teams often deploy these tools as isolated assistants. The resulting handoffs separate requirements, architecture, implementation, security, testing, deployment, and operational evidence. This paper presents the Enterprise Multi-Agent Software Engineering (EMASE) framework, a conceptual architecture for coordinating specialized agents across that lifecycle. EMASE assigns agents to requirements analysis, enterprise knowledge retrieval, solution architecture, code generation, security review, testing, deployment, production monitoring, and organizational learning. An orchestration engine controls task order, artifact exchange, policy checks, conflict handling, and human approval. A shared knowledge layer preserves traceable engineering artifacts and operational feedback. A representative enterprise scenario shows how a requirement moves through the agents and where human decisions remain mandatory. The paper does not report a production experiment; instead, it defines an implementable architecture, qualitative comparison, and evaluation plan for studying coordination quality, software quality, traceability, and operational reliability.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.