MASAI: Modular Architecture for Software-engineering AI Agents
This addresses the challenge of automating software engineering tasks for developers, representing a strong specific gain rather than a broad breakthrough.
The paper tackled the problem of solving complex software engineering tasks by proposing MASAI, a modular architecture with LLM-powered sub-agents, achieving a 28.33% resolution rate on the SWE-bench Lite dataset.
A common method to solve complex problems in software engineering, is to divide the problem into multiple sub-problems. Inspired by this, we propose a Modular Architecture for Software-engineering AI (MASAI) agents, where different LLM-powered sub-agents are instantiated with well-defined objectives and strategies tuned to achieve those objectives. Our modular architecture offers several advantages: (1) employing and tuning different problem-solving strategies across sub-agents, (2) enabling sub-agents to gather information from different sources scattered throughout a repository, and (3) avoiding unnecessarily long trajectories which inflate costs and add extraneous context. MASAI enabled us to achieve the highest performance (28.33% resolution rate) on the popular and highly challenging SWE-bench Lite dataset consisting of 300 GitHub issues from 11 Python repositories. We conduct a comprehensive evaluation of MASAI relative to other agentic methods and analyze the effects of our design decisions and their contribution to the success of MASAI.