Each developer participated in the test group for half of the tasks and in the control group for the other half. Each task was performed by a test group that had access to two generative AI–based tools and a control group that used no AI assistance. This lab will serve as an ongoing test bed to understand developments in the industry, including the impact of new tools and developments in existing tools.įor this report, participants were asked to perform common software development tasks in three areas-code generation, refactoring, and documentation-over the course of several weeks. To understand the impact of generative AI–based tools on developer productivity, we set up a lab with more than 40 McKinsey developers who are located across the United States and Asia and have different amounts of software development experience. With the right upskilling and enterprise enablers, these speed gains can be translated into an increase in productivity that outperforms past advances in engineering productivity, driven by both new tooling and processes. Documenting code functionality for maintainability (which considers how easily code can be improved) can be completed in half the time, writing new code in nearly half the time, and optimizing existing code (called code refactoring) in nearly two-thirds the time (Exhibit 1). delivering impressive speed gains for many common developer tasks (see sidebar, “About the research”). Our latest empirical research finds generative AI–based tools 1 Includes both generative AI-based tools trained to have natural conversations through prompting and those trained specifically on code base and embedded into a developer’s integrated development environment (IDE).
This article is a collaborative effort by Begum Karaci Deniz, Chandra Gnanasambandam, Martin Harrysson, Alharith Hussin, and Shivam Srivastava, representing views from McKinsey Digital.
0 kommentar(er)
