Your Code Agent Can Grow Alongside You with Structured Memory
This addresses the challenge of improving code agents' adaptability and performance in complex software engineering for developers, representing a novel method rather than an incremental improvement.
The paper tackles the problem of code agents being limited by static code snapshots, which hinders their ability to handle complex software engineering tasks, by proposing MemCoder, a framework that enables continual human-AI co-evolution through structured memory and self-refinement; it achieves state-of-the-art performance with a 9.4% improvement in resolved rate over DeepSeek-V3.2 on SWE-bench Verified.
While "Intent-oriented programming" (or "Vibe Coding") redefines software engineering, existing code agents remain tethered to static code snapshots. Consequently, they struggle to model the critical information embedded in the temporal evolution of projects, failing to leverage the "reasoning trajectories" implicit in past successful practices. This limitation results in rigid behavioral logic and a lack of autonomous adaptability, ultimately hindering their ability to tackle complex, repository-level problems. To bridge this static-dynamic mismatch, we propose MemCoder, a framework designed to enable continual human-AI co-evolution. MemCoder first structures historical human experience to distill latent intent-to-code mappings from past commits. It then employs a self-refinement mechanism driven by verification feedback to correct agent behavior in real-time. Crucially, an experience self-internalization mechanism is introduced to crystallize human-validated solutions into long-term knowledge, thereby supporting sustained evolution. Experimental results on SWE-bench Verified demonstrate that MemCoder not only achieves State-of-the-Art (SOTA) performance but also delivers a 9.4% improvement in resolved rate over the general foundation model DeepSeek-V3.2. These findings indicate that equipping agents with the capability to co-evolve with humans via project history and real-time feedback effectively unlocks the potential of general models in complex software engineering tasks.