Self-Organizing Multi-Agent Systems for Continuous Software Development

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.

Figures (2)

• Figure 1: Architecture of TheBotCompany. The human monitors the system through a dashboard connected to the orchestrator, which drives a three-phase milestone lifecycle (Strategy $\to$ Execution $\to$ Verification). Manager agents govern each phase and dynamically hire/fire worker agents. Solid arrows show normal progression; dashed arrows show recovery paths. All agents access shared infrastructure through a common bus.
• Figure 2: The monitoring dashboard showing project status, orchestrator state, cost tracking, worker activity, open PRs, and live agent output streaming.