Paradigm shift on Coding Productivity Using GenAI
Liang Yu
TL;DR
This study investigates industrial adoption of GenAI coding assistants within FinTech and telecom settings to quantify productivity impacts and extract practical lessons. Using a two-phase, multi-case design with surveys and interviews (n=18), the authors identify five key productivity factors and show that GenAI delivers clear benefits for routine tasks but limited gains for complex, domain-specific work due to context gaps and lack of custom design-rule support. Iterative prompt refinement and immersive IDE use emerge as important practices to improve outcomes, while automated quality evaluation remains a critical unmet need. The findings highlight that domain knowledge and tool integration shape GenAI value, suggesting future work should expand to additional domains and develop context-aware, rule-driven GenAI capabilities to realize broader productivity gains.
Abstract
Generative AI (GenAI) applications are transforming software engineering by enabling automated code co-creation. However, empirical evidence on GenAI's productivity effects in industrial settings remains limited. This paper investigates the adoption of GenAI coding assistants (e.g., Codeium, Amazon Q) within telecommunications and FinTech domains. Through surveys and interviews with industrial domain-experts, we identify primary productivity-influencing factors, including task complexity, coding skills, domain knowledge, and GenAI integration. Our findings indicate that GenAI tools enhance productivity in routine coding tasks (e.g., refactoring and Javadoc generation) but face challenges in complex, domain-specific activities due to limited context-awareness of codebases and insufficient support for customized design rules. We highlight new paradigms for coding transfer, emphasizing iterative prompt refinement, immersive development environment, and automated code evaluation as essential for effective GenAI usage.