[4Xin2] Poster session 2

[4Xin2-39] Preliminary Investigation of Essential Skills for Programming with Generative AI

〇Miyu Sato¹, Yui Noguchi¹, Waka Ito¹, Teruno Kajiura¹, Shiho Takano¹, Kimio Kuramitsu¹ (1.Japan Women's University)

Keywords:Generative AI, Programming

Generative AI is beginning to have a significant impact on software development and programming education.This impact raises questions about the validity of traditional programming skills and the need to learn new skills.Understanding the essential skills for programming with generative AI is important for shaping the future of software development and programming education. This study investigates how differences in developer skill levels affect the quality of code generated by generative AI. We classify the skills into four skills: coding, software process, domain knowledge, and prompt engineering, and assess them using three code generation tasks.Preliminary investigation suggests that prompt engineering could influence the quality of code generated by AI. This presentation reports on these initial findings and seeks to discuss the prospects for more detailed research into the skills required for effective use of generative AI.