A Survey of Vibe Coding with Large Language Models
🤖 AI Summary
“Vibe Coding”—a novel human-AI collaborative programming paradigm relying on outcome-based feedback rather than code comprehension—suffers from unquantified productivity losses and collaboration bottlenecks.
Method: We formalize Vibe Coding as a constrained Markov Decision Process (MDP), propose a unified taxonomy encompassing five developer modeling paradigms, and design an integrated technical ecosystem comprising coding agents, context-enhanced development environments, and dynamic feedback mechanisms. Our evaluation combines systematic literature review (1,000+ papers), MDP-theoretic analysis, LLM-based programming experiments, and test-driven validation.
Contribution/Results: This work establishes the first theoretical foundation and practical framework for Vibe Coding. We identify context engineering and human-AI collaboration patterns as primary determinants of efficiency, and provide a scalable, empirically grounded methodology for human-AI co-development in software engineering.
Application Category
📝 Abstract
The advancement of large language models (LLMs) has catalyzed a paradigm shift from code generation assistance to autonomous coding agents, enabling a novel development methodology termed "Vibe Coding" where developers validate AI-generated implementations through outcome observation rather than line-by-line code comprehension. Despite its transformative potential, the effectiveness of this emergent paradigm remains under-explored, with empirical evidence revealing unexpected productivity losses and fundamental challenges in human-AI collaboration. To address this gap, this survey provides the first comprehensive and systematic review of Vibe Coding with large language models, establishing both theoretical foundations and practical frameworks for this transformative development approach. Drawing from systematic analysis of over 1000 research papers, we survey the entire vibe coding ecosystem, examining critical infrastructure components including LLMs for coding, LLM-based coding agent, development environment of coding agent, and feedback mechanisms. We first introduce Vibe Coding as a formal discipline by formalizing it through a Constrained Markov Decision Process that captures the dynamic triadic relationship among human developers, software projects, and coding agents. Building upon this theoretical foundation, we then synthesize existing practices into five distinct development models: Unconstrained Automation, Iterative Conversational Collaboration, Planning-Driven, Test-Driven, and Context-Enhanced Models, thus providing the first comprehensive taxonomy in this domain. Critically, our analysis reveals that successful Vibe Coding depends not merely on agent capabilities but on systematic context engineering, well-established development environments, and human-agent collaborative development models.
Problem
Research questions and friction points this paper is trying to address.
Surveying Vibe Coding paradigm with large language models
Investigating productivity challenges in human-AI coding collaboration
Establishing theoretical foundations for autonomous coding agents
Innovation
Methods, ideas, or system contributions that make the work stand out.
Constrained Markov Decision Process formalizes Vibe Coding
Synthesizes five distinct development models for taxonomy
Systematic context engineering enhances human-agent collaboration
🔎 Similar Papers
No similar papers found.
