Paper Type

Short

Paper Number

PACIS2025-1336

Description

This paper aims to understand the interaction between developers and Generative Artificial Intelligence (GenAI) using a workflow perspective. We interviewed 19 developers who use GenAI tools in their daily work. Through preliminary analysis, we identified three distinct developer-AI workflows and the work activities where GenAI assists within each workflow. We visualized these workflows taking conversational assistance type GenAI as an example and explained the decision points and iterative interactions that characterize developer-GenAI collaboration. By shifting the focus from individual tasks to the broader workflow, this research provides a comprehensive understanding of how GenAI integrates into developers’ work in a real-world context. The findings offer valuable insights into the evolving dynamics of human-AI collaboration and inform the design of more effective and sustainable human-AI workflows in the context of software development.

Comments

Design

Share

COinS
 
Jul 6th, 12:00 AM

Developer-GenAI Collaboration: Understanding Workflow Dynamics for Sustainable Human-AI Interaction

This paper aims to understand the interaction between developers and Generative Artificial Intelligence (GenAI) using a workflow perspective. We interviewed 19 developers who use GenAI tools in their daily work. Through preliminary analysis, we identified three distinct developer-AI workflows and the work activities where GenAI assists within each workflow. We visualized these workflows taking conversational assistance type GenAI as an example and explained the decision points and iterative interactions that characterize developer-GenAI collaboration. By shifting the focus from individual tasks to the broader workflow, this research provides a comprehensive understanding of how GenAI integrates into developers’ work in a real-world context. The findings offer valuable insights into the evolving dynamics of human-AI collaboration and inform the design of more effective and sustainable human-AI workflows in the context of software development.