Problem
Research questions and friction points this paper is trying to address.
continuous software development
multi-agent systems
large language models
long-term development
software automation
Self-Organizing Multi-Agent Systems for Continuous Software Development
🤖 AI Summary
This work addresses the challenge that existing large language model–based multi-agent systems struggle to support sustained, continuous software development. The authors propose TheBotCompany, a novel framework that orchestrates milestone-driven development through a three-stage state machine—planning, execution, and verification—and dynamically forms self-organizing agent teams by hiring and releasing agents on demand. An asynchronous human oversight mechanism is integrated to ensure alignment with project goals. This approach achieves, for the first time, multi-day, end-to-end automated software development in real-world projects, significantly improving milestone completion rates and code quality while effectively identifying latent defects—all without compromising cost efficiency.
📝 Abstract
Large Language Model-based multi-agent systems have shown promise in automating software development tasks. However, most vibe code systems focus on completing small tasks and incremental code changes, leaving persistent, continuous software development largely unexplored. We present TheBotCompany, an open-source orchestration framework for continuous multi-agent software development. TheBotCompany introduces three key innovations: (1) a three-phase state machine (Strategy to Execution to Verification) for milestone-driven development, (2) self-organizing agent teams where manager agents dynamically hire, assign, and fire worker agents based on project needs, and (3) asynchronous human oversight. We evaluate TheBotCompany on real-world software projects over multiple days of continuous development, measuring team adaptation patterns, milestone completion rates, cost efficiency, and code quality. Our results demonstrate that the self-organizing approach enables effective long-term software development with measurable progress, while the verification phase catches defects that would otherwise persist.
Innovation
Methods, ideas, or system contributions that make the work stand out.
self-organizing multi-agent systems
continuous software development
LLM-based agents
milestone-driven development
asynchronous human oversight
🔎 Similar Papers
No similar papers found.
