Vibe coding: programming through conversation with artificial intelligence

2025-06-29

AI Summary

This study addresses “ambient programming”—an emerging paradigm wherein developers construct software primarily through conversational interaction with large language models (LLMs) rather than manual coding—raising critical questions about goal formulation, prompting strategies, debugging mechanisms, and foundational challenges. Method: Drawing on think-aloud programming session videos, we conduct a qualitative study using framework analysis to systematically examine practitioner practices. Contribution/Results: We introduce and explicate two key constructs: (1) a dual-loop iterative mechanism—comprising goal setting, validation, and revision—and (2) “material detachment,” a phenomenon reflecting epistemic distance from code artifacts. Findings reveal a shift in programming expertise from syntactic implementation toward contextual modeling, real-time trust calibration, and human-AI co-decision-making. Trust emerges as a dynamic, evidence-supported process; AI does not supplant expertise but reconfigures its distribution and enactment, establishing a novel human-AI collaborative programming paradigm.

Technology Category

• Reasoning under Uncertainty: Probabilistic Programming
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

We examine "vibe coding": an emergent programming paradigm where developers primarily write code by interacting with code-generating large language models rather than writing code directly. We analysed a curated set of videos depicting extended vibe coding sessions with rich think-aloud reflections. Using framework analysis, we investigated programmers' goals, workflows, prompting techniques, debugging approaches, and challenges encountered. We find that vibe coding follows iterative goal satisfaction cycles where developers alternate between prompting AI, evaluating generated code through rapid scanning and application testing, and manual editing. Prompting strategies blend vague, high-level directives with detailed technical specifications. Debugging remains a hybrid process combining AI assistance with manual practices. Critically, vibe coding does not eliminate the need for programming expertise but rather redistributes it toward context management, rapid code evaluation, and decisions about when to transition between AI-driven and manual manipulation of code. Trust in AI tools during vibe coding is dynamic and contextual, developed through iterative verification rather than blanket acceptance. Vibe coding is an evolution of AI-assisted programming that represents an early manifestation of "material disengagement", where practitioners orchestrate code production and manipulation, mediated through AI, while maintaining selective and strategic oversight.

Problem

• Studying how developers write code using AI-generated code
• Analyzing programmers' goals, workflows, and challenges with AI coding
• Investigating how trust and expertise evolve in AI-assisted programming

Innovation

• Iterative goal satisfaction cycles with AI
• Hybrid debugging combining AI and manual practices
• Dynamic trust through iterative verification