Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AI

Ranjan Sapkota 1 3, Konstantinos I. Roumeliotis 2, Manoj Karkee 1 3 · 2025

Read Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AI by Ranjan Sapkota 1 3, Konstantinos I. Roumeliotis 2, Manoj Karkee 1 3 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-19443.txt.

Introduction

Ranjan Sapkota, Konstantinos Roumeliotis, and Manoj Karkee compare two styles of AI-assisted coding in a review from two thousand twenty-five. Both lean on large language models, but they split the work between human and machine very differently. Vibe coding keeps the human in a chatty loop; agentic coding hands whole goals to autonomous agents.

The review builds a taxonomy of design, feedback, safety, and debugging across the two styles. Twenty use cases show vibe tools shining in prototypes and teaching, while agents excel at enterprise automation and pipeline work. Hybrid designs that pair chat interfaces with agent back ends are rising. The authors predict success lies in blending both, inside a human-centered process, rather than picking a winner.

Catalog record

CREATORRanjan Sapkota 1 3, Konstantinos I. Roumeliotis 2, Manoj Karkee 1 3
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE121 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-19443

Browse the full library · About the Tower of Babel