Building Software by Rolling the Dice: A Qualitative Study of Vibe Coding

• URL: http://arxiv.org/abs/2512.22418v2
• Date: Tue, 30 Dec 2025 20:19:49 GMT
• Title: Building Software by Rolling the Dice: A Qualitative Study of Vibe Coding
• Authors: Yi-Hung Chou, Boyuan Jiang, Yi Wen Chen, Mingyue Weng, Victoria Jackson, Thomas Zimmermann, James A. Jones,
• Abstract summary: "vibe coders" build software primarily through prompts rather than writing code.<n>We conducted a theory study of 20 vibe-coding videos, including 7 live-streamed coding sessions and 13 opinion videos.<n>Our findings reveal a spectrum of behaviors: some vibe coders rely almost entirely on AI, while others examine and adapt generated outputs.
• Score: 15.145249560710377
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) are reshaping software engineering by enabling "vibe coding," in which developers build software primarily through prompts rather than writing code. Although widely publicized as a productivity breakthrough, little is known about how practitioners actually define and engage in these practices. To shed light on this emerging phenomenon, we conducted a grounded theory study of 20 vibe-coding videos, including 7 live-streamed coding sessions (about 16 hours, 254 prompts) and 13 opinion videos (about 5 hours), supported by additional analysis of activity durations and prompt intents. Our findings reveal a spectrum of behaviors: some vibe coders rely almost entirely on AI without inspecting code, while others examine and adapt generated outputs. Across approaches, all must contend with the stochastic nature of generation, with debugging and refinement often described as "rolling the dice." Further, divergent mental models, shaped by vibe coders' expertise and reliance on AI, influence prompting strategies, evaluation practices, and levels of trust. These findings open new directions for research on the future of software engineering and point to practical opportunities for tool design and education.

Related papers

• A Grounded Theory of Debugging in Professional Software Engineering Practice [5.88596290266904]
  We conducted a study using a grounded theory approach.<n>We observed seven professional developers and five professional live-coding streamers working on 17 debug tasks.
  arXiv  Detail & Related papers  (2026-02-11T23:19:45Z)
• Guidelines to Prompt Large Language Models for Code Generation: An Empirical Characterization [82.29178197694819]
  We derive and evaluate development-specific prompt optimization guidelines.<n>We use an iterative, test-driven approach to automatically refine code generation prompts.<n>We conduct an assessment with 50 practitioners, who report their usage of the elicited prompt improvement patterns.
  arXiv  Detail & Related papers  (2026-01-19T15:01:42Z)
• "Can you feel the vibes?": An exploration of novice programmer engagement with vibe coding [42.82674998306379]
  "vibe coding" refers to creating software via natural language prompts rather than direct code authorship.<n>This paper reports on a one-day educational hackathon investigating how novice programmers and mixed-experience teams engage with vibe coding.
  arXiv  Detail & Related papers  (2025-12-02T13:32:23Z)
• From Code Foundation Models to Agents and Applications: A Practical Guide to Code Intelligence [150.3696990310269]
  Large language models (LLMs) have transformed automated software development by enabling direct translation of natural language descriptions into functional code.<n>We provide a comprehensive synthesis and practical guide (a series of analytic and probing experiments) about code LLMs.<n>We analyze the code capability of the general LLMs (GPT-4, Claude, LLaMA) and code-specialized LLMs (StarCoder, Code LLaMA, DeepSeek-Coder, and QwenCoder)
  arXiv  Detail & Related papers  (2025-11-23T17:09:34Z)
• Rapid Development of Omics Data Analysis Applications through Vibe Coding [0.0]
  I demonstrate that modern large language models (LLMs) and autonomous coding agents can dramatically lower this barrier.<n>I used Vibe coding to create a fully functional data analysis website capable of performing standard tasks.<n>The entire application, including user interface, backend logic, and data upload pipeline, was developed in less than ten minutes using only four natural-language prompts.
  arXiv  Detail & Related papers  (2025-10-10T19:06:27Z)
• Prompting in Practice: Investigating Software Developers' Use of Generative AI Tools [17.926187565860232]
  The integration of generative artificial intelligence (GenAI) tools has fundamentally transformed software development.<n>This study presents a systematic investigation of how software engineers integrate GenAI tools into their professional practice.<n>We surveyed 91 software engineers, including 72 active GenAI users, to understand AI usage patterns throughout the development process.
  arXiv  Detail & Related papers  (2025-10-07T15:02:22Z)
• Modeling Student Learning with 3.8 Million Program Traces [52.153493498021895]
  We introduce a dataset of over 3.8 million programming reasoning traces from users of Pencil Code.<n>We find that models trained on real traces are stronger at modeling diverse student behavior.<n>We show that we can help students recover from mistakes by steering code generation models to identify a sequence of edits that will result in more correct code.
  arXiv  Detail & Related papers  (2025-10-06T17:37:17Z)
• Vibe Coding in Practice: Motivations, Challenges, and a Future Outlook -- a Grey Literature Review [2.5195922470930614]
  Vibe coding is the practice where users rely on AI code generation tools through intuition and trial-and-error without necessarily understanding the underlying code.<n>No research has systematically investigated why users engage in vibe coding, what they experience while doing so, and how they approach quality assurance (QA) and perceive the quality of the AI-generated code.<n>Our analysis reveals a speed-quality trade-off paradox, where vibe coders are motivated by speed and accessibility, often experiencing rapid instant success and flow'', yet most perceive the resulting code as fast but flawed.
  arXiv  Detail & Related papers  (2025-09-30T22:35:00Z)
• Good Vibrations? A Qualitative Study of Co-Creation, Communication, Flow, and Trust in Vibe Coding [6.862249355928346]
  We propose a grounded theory of vibe coding centered on conversational interaction with AI, co-creation, and developer flow and joy.<n>We find that AI trust regulates movement along a continuum from delegation to co-creation and supports the developer experience by sustaining flow.
  arXiv  Detail & Related papers  (2025-09-15T22:28:42Z)
• How Far Have We Gone in Binary Code Understanding Using Large Language Models [51.527805834378974]
  We propose a benchmark to evaluate the effectiveness of Large Language Models (LLMs) in binary code understanding. Our evaluations reveal that existing LLMs can understand binary code to a certain extent, thereby improving the efficiency of binary code analysis.
  arXiv  Detail & Related papers  (2024-04-15T14:44:08Z)
• Collaborative, Code-Proximal Dynamic Software Visualization within Code Editors [55.57032418885258]
  This paper introduces the design and proof-of-concept implementation for a software visualization approach that can be embedded into code editors. Our contribution differs from related work in that we use dynamic analysis of a software system's runtime behavior. Our visualization approach enhances common remote pair programming tools and is collaboratively usable by employing shared code cities.
  arXiv  Detail & Related papers  (2023-08-30T06:35:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.