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

TL;DR

EvoDev targets the inefficiencies of linear, end-to-end software development with LLM agents by introducing a feature-driven, iterative framework that explicitly models dependencies via a Feature Map DAG and propagates multi-layer context across iterations. Built on an FDD-inspired workflow, EvoDev uses dedicated agents for overall design, feature extraction, planning, and implementation, and incorporates a memory-augmented pipeline to reduce context bloat. Empirical evaluation on Android development tasks shows EvoDev outperforms baselines (e.g., Claude Code) with a $56.8\%$ improvement in functional completeness and up to $76.6\%$ gains in certain models, while maintaining competitive costs. The work offers practical design insights for iterative AI-assisted software engineering and highlights how dependency modeling and context propagation can enable scalable, robust end-to-end development with LLMs.

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.

Figures (6)

• Figure 1: Overview of the FDD-inspired EvoDev framework
• Figure 2: The basic activities of Feature Driven Development
• Figure 3: The requirement document for the countdown timer APP
• Figure 4: The overall design (including both UI and data) for the countdown timer APP
• Figure 5: The feature list for the countdown timer APP. The bold red highlights indicate the UI and data designs referenced in the UI flow and data flow descriptions.