EvoDev: An Iterative Feature-Driven Framework for End-to-End Software Development with LLM-based Agents

2025-11-04

AI Summary

Current large language model (LLM)-based agents typically adopt linear, waterfall-style pipelines for end-to-end software development, failing to accommodate the iterative nature and complex interdependencies inherent in real-world development. To address this, we propose EvoDev—a feature-driven, iterative framework inspired by Feature-Driven Development (FDD). EvoDev first models natural language requirements as directed acyclic feature graphs with explicit functional dependencies. It then introduces a multi-level information propagation mechanism enabling cross-turn context retention and workflow-aware agent collaboration. This breaks the unidirectional pipeline paradigm, establishing a closed-loop iteration over requirement decomposition, incremental implementation, and dependency coordination. Evaluated on Android development tasks, EvoDev achieves a 56.8% improvement over Claude Code. Across diverse base LLMs, single-agent performance improves by 16.0%–76.6%, demonstrating both generality and effectiveness.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Modeling other Agents

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: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Recent advances in large language model agents offer the promise of automating end-to-end software development from natural language requirements. However, existing approaches largely adopt linear, waterfall-style pipelines, which oversimplify the iterative nature of real-world development and struggle with complex, large-scale projects. To address these limitations, we propose EvoDev, an iterative software development framework inspired by feature-driven development. EvoDev decomposes user requirements into a set of user-valued features and constructs a Feature Map, a directed acyclic graph that explicitly models dependencies between features. Each node in the feature map maintains multi-level information, including business logic, design, and code, which is propagated along dependencies to provide context for subsequent development iterations. We evaluate EvoDev on challenging Android development tasks and show that it outperforms the best-performing baseline, Claude Code, by a substantial margin of 56.8%, while improving single-agent performance by 16.0%-76.6% across different base LLMs, highlighting the importance of dependency modeling, context propagation, and workflow-aware agent design for complex software projects. Our work summarizes practical insights for designing iterative, LLM-driven development frameworks and informs future training of base LLMs to better support iterative software development.

Problem

• Addresses limitations of linear development in LLM-based software automation
• Models feature dependencies through directed graphs for iterative development
• Enhances context propagation across business logic, design, and code

Innovation

• Iterative framework with feature-driven development approach
• Feature Map models dependencies as directed acyclic graph
• Multi-level context propagation along feature dependencies