Adlene Sifi explores the impact of generative AI on developer experience.

In this article, we will try to determine if there is a link between the use of generative AI (e.g., GitHub Copilot) and developer experience (DevEx). Specifically, we aim to verify whether the use of generative AI has a positive impact on developer experience. We will even try to see if there is a causal link between the use of generative AI and the improvement of developer experience.

First, let’s define these two concepts.

What is developer experience? Is it a new buzzword? A market trend? Worse, is it a new marketing tactic to make us buy more products? No, it is none of these. Otherwise, I would never have considered writing this article. Developer experience is something more fascinating, more intriguing, but above all, more holistic. Here are some definitions:

• “Developer experience encompasses how developers feel about, think about, and value their work” (Nicole, Margaret-Anne, Abi, & Michaela, 2023)
• “Developer experience (DevEx) refers to all aspects of interactions between developers and the tools, platforms, processes, and people they work with to deliver software products and services” (Gartner, 2024)
• “DevEx refers to the systems, technology, process, and culture that influence the effectiveness of software development. It looks at all components of a developer’s ecosystem—from environment to workflows to tools—and asks how they are contributing to developer productivity, satisfaction, and operational impact” (Davis, 2023)

These definitions speak of the experience lived by a software developer when designing a product. They also mention the factors that influence this experience. Among these factors are: the work culture in which the software developer operates, the colleagues and teams they collaborate with, the processes governing their actions, and the tools aiding their daily work. These factors can influence the developer experience either positively or negatively. Ultimately, there seems to be a direct link between developer experience and the productivity of software developers.

Imagine for a moment that your software developer has the best tools on the market. However, they are immersed in a company culture that does not encourage questioning the status quo, where divergent views are not tolerated, and where criticism is met with hostility. This culture of fear will clearly negatively impact the developer experience. Do you believe that under these conditions, your software developer will be highly productive? Do you think they will thrive at work? This is highly unlikely.

Here is another very common scenario: Your software developers have the best tools on the market. Their work processes, although undocumented, are still quite good. However, inter-team collaboration is not encouraged at all, and any work dependent on two or more teams takes forever. Does this sound familiar? Do you believe that under these conditions, your software developers will be highly productive? Once again, this is highly unlikely.

It only takes one defective factor of the developer experience for the entire structure to collapse and for your overall productivity to suffer.

The developer experience is like a house of cards. If one of the cards (culture, processes, collaboration, tools) is missing, the entire structure wobbles and may fall. Therefore, it is essential to pay great attention to each factor that directly impacts the developer experience.

“There is truly a difference between writing code and writing code in an optimized environment for coding. Optimized coding environments are efficient, effective, conducive to well-being, and depend on the right balance between technologies, processes, and social structures” (Nicole, Eirini, Abi, Michaela, & Brian, 2024)

How is this developer experience useful? What value does it bring to organizations?

According to a McKinsey study conducted in 2020, organizations that offer a good work environment to their software developers had revenues 4 to 5 times greater than those of their competition (Nicole, Margaret-Anne, Abi, & Michaela, 2023).

According to Gartner, organizations that offer superior developer experience to their software developers are (Gartner, 2024):

• 33% more likely to achieve their business goals
• 31% more likely to improve their delivery chain
• 20% more likely to retain their employees

In a McKinsey study, a major SaaS solution provider saw the productivity of its 2000 developers increase by 20% and a reduction in vulnerabilities by 15% to 20% by implementing a developer experience program (Thomas, Arun, Ling, Stephan, & Lars, 2022).

Finally, a study conducted on 1000 retail stores in the United States showed that stores offering a better employee experience could increase their revenues by 50% (Kate, Tiffani, Kexin, & Lalith, 2022). Even though this talks about employee experience and not specifically developer experience, there is a positive correlation between these two types of experiences.

Now that we have a better understanding of what developer experience is and its impacts, let’s briefly describe what generative AI is. Here are some definitions:

• “Generative AI can be thought of as a machine-learning model that is trained to create new data, rather than making a prediction about a specific dataset. A generative AI system is one that learns to generate more objects that look like the data it was trained on” (Zewe, 2023)
• “Generative artificial intelligence (AI) describes algorithms (such as ChatGPT) that can be used to create new content, including audio, code, images, text, simulations, and videos. Recent breakthroughs in the field have the potential to drastically change the way we approach content creation” (McKinsey, 2024)

According to these definitions, generative AI, and more specifically the Transformer model, is a deep learning model that allows creating different types of content (multimodal): text, image, voice, or video. This model stands out from traditional artificial intelligence models, which were mainly designed for either classification (e.g.: identifying a dog among a set of animals) or regression (e.g.: predicting the price of a house).

We have just defined what developer experience (DevEx) is and demonstrated its impact on profitability. We have also explained what generative AI is. Now let’s see how the latter can improve developer experience (DevEx). But before getting there, let’s first explain what it takes to improve DevEx.

We have seen earlier that DevEx is influenced by company culture, work processes, and work tools. As you can imagine, these three axes are very vast, and trying to intervene on them is a laborious process that often occurs during an organizational transformation. However, it is possible to influence DevEx by specifically acting on dimensions with a reduced scale. These dimensions are: the feedback loop, cognitive load, and the flow state (Nicole, Margaret-Anne, Abi, & Michaela, 2023).

The feedback loop occurs when the output of one system becomes the input of another system. More specifically for a software developer, this happens when the team’s feedback (output) in the form of a code review or a response to one of their questions influences (input) the changes they made to the system. In the context of DevEx, what interests us more specifically is the speed with which the developer receives this feedback from their team. Software developers who received quick feedback from their team reported being 20% more innovative than those who reported longer feedback delays. Meanwhile, teams providing quick feedback to their developers reported having 50% less technical debt than teams whose feedback took longer (Nicole, Eirini, Abi, Michaela, & Brian, 2024).

Cognitive load corresponds to the amount of information that working memory can store and process simultaneously in a short period. In the context of DevEx, cognitive load corresponds to the mental effort needed by a software developer to accomplish a task. Cognitive load tends to increase when the developer works on complex tasks or when they try to understand new technology. Cognitive load also varies with the way information is presented or structured. Software developers who reported a good understanding of the source code they work with said they were 42% more productive than those who had limited understanding of the source code. Moreover, developers who find their work tools user-friendly and easy to use feel 50% more innovative than those who find their tools more complex and less user-friendly (Nicole, Eirini, Abi, Michaela, & Brian, 2024).

Finally, the flow state is a mental state in which a person is fully immersed in their work. It is a state of absolute concentration. In the context of DevEx, the flow state is measured by the level of satisfaction with the time dedicated to accomplishing deep work without interruption. Software developers who have sufficient time for deep work reported being 50% more productive than those who have less time dedicated to deep work. Additionally, developers who find their work more engaging said they were 30% more productive than those who find their work boring or uninteresting (Nicole, Eirini, Abi, Michaela, & Brian, 2024).

Now that we have explained the three dimensions that allow improving DevEx on a smaller scale than company culture or organizational processes, let’s see how generative AI can improve DevEx by directly influencing these dimensions.

Let’s start with the feedback loop. A study conducted by Accenture with its developers showed that GitHub Copilot increased the approval rate of changes made by software developers by 15% (Gao & GitHub, 2024). Another study demonstrated that change requests (Pull Request) described using GitHub Copilot have a 13% higher approval rate than change requests (Pull Request) described by the software developer. This same study showed that using GitHub Copilot for describing change requests reduces code review (feedback) time by 19.3 hours and increases the chances of approval of change requests by 1.57 times (Xiao, Hata, Treude, & Matsumoto, 2024).

I have supported many organizations and thousands of software developers in adopting GitHub Copilot. I would add that the introduction of the code review feature in GitHub Copilot has significantly reduced the time required for teams to approve changes. Software developers use it to ensure that their changes comply with their team’s programming standards, and teams use it to quickly validate the conformity of changes introduced by team members.

Regarding cognitive load, a McKinsey study showed that generative AI reduces the development time of complex tasks by 12% and that of medium tasks by 10% (McKinsey, 2023). A survey conducted with 2000 software developers worldwide showed that 60%-71% of respondents find it easy to learn a new programming language or understand existing code thanks to generative AI, while 23%-29% of respondents find the same tasks very easy to achieve thanks to generative AI (Daigle & GitHub, 2024). Finally, another study showed that 70% of software developers use less mental load with repetitive tasks thanks to GitHub Copilot, while 54% of them spend less time searching for information (Gao & GitHub, 2024).

Regarding the impact of generative AI and specifically GitHub Copilot on cognitive load among my clients, I have observed a new phenomenon: several roles that did not necessarily involve programming before have started writing code due to the significant reduction in cognitive load. Writing code has become accessible to everyone! This democratizes software development and creates new opportunities for organizations.

Lastly, several studies have demonstrated the impact of generative AI on the flow state. A McKinsey study showed that software developers using generative AI felt 39% more in the flow state than those who did not use it (McKinsey, 2023). Another study showed that GitHub Copilot users were 22% more focused on satisfying work than those who did not use it (Wivestad, Stray, & Barbala, 2024).

Once again, I have observed the impact of GitHub Copilot on the flow state among many of my customers. The ability of GitHub Copilot to let the developer interact with various information sources without leaving the development environment (IDE) has a real impact on concentration and significantly limits context switching. Before the advent of generative AI, software developers often had to switch contexts to seek information from various websites such as Stack Overflow or Microsoft Learn. Now, thanks to generative AI, it is possible to access these information sources directly from the development environment.

Finally, we have arrived! In this article, we first defined the concepts of developer experience (DevEx) and generative AI. Next, we examined the dimensions of DevEx. Lastly, we explored how generative AI influences DevEx. The data presented indicates a clear connection between generative AI and developer experience. It suggests that the use of generative AI has a notable impact on developer experience. What are your thoughts?

References

Daigle, K., & GitHub. (2024, September). Survey: The AI wave continues to grow on software development teams. Retrieved from github.blog: https://github.blog/news-insights/research/survey-ai-wave-grows/

Davis, G. (2023, June). Developer experience: What is it and why should you care? Retrieved from GitHub.blog: https://github.blog/enterprise-software/collaboration/developer-experience-what-is-it-and-why-should-you-care/

Gao, Y., & GitHub. (2024, May 13). github.blog. Retrieved from Research: Quantifying GitHub Copilot’s impact in the enterprise with Accenture: https://github.blog/news-insights/research/research-quantifying-github-copilots-impact-in-the-enterprise-with-accenture/

Gartner. (2024). Developer Experience (DevEx) as a Key Driver of Productivity. Retrieved from Gartner.com: https://www.gartner.com/en/software-engineering/topics/developer-experience

Kate, G., Tiffani, B., Kexin, C., & Lalith, M. (2022, March). Research: How Employee Experience Impacts Your Bottom Line. Retrieved from hbr.org: https://hbr.org/2022/03/research-how-employee-experience-impacts-your-bottom-line

Keith, O., Amanda, D., & Molly, H. (2023). What is employee experience? Retrieved from ibm.com: https://www.ibm.com/topics/employee-experience

Langerock, N., Oberauer, K., Throm, E., & Vergauwe, E. (2025). The cognitive load effect in working memory: Refreshing the empirical landscape, removing outdated explanations. ScienceDirect.

McKinsey. (2023, June). Unleashing developer productivity with generative AI. Retrieved from mckinsey.com: https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/unleashing-developer-productivity-with-generative-ai

McKinsey. (2024, 4 2). What is generative AI? Retrieved from www.mckinsey.com: https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai

Nicole, F., Eirini, K., Abi, N., Michaela, G., & Brian, H. M.-A. (2024). DevEx in Action. ACM, 30.

Nicole, F., Margaret-Anne , S., Abi, N., & Michaela , G. (2023). DevEx: What Actually Drives Productivity. acmqueue, 19.

Thomas, D., Arun, G., Ling, L., Stephan, S., & Lars, S. (2022, June). Why you should be thinking about DX. Retrieved from mckinsey.com: https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/tech-forward/why-your-it-organization-should-prioritize-developer-experience

Wivestad, V. T., Stray, V., & Barbala, A. M. (2024). Copilot’s Island of Joy: Balancing Individual Satisfaction with Team Interaction in Agile Development.

Xiao, T., Hata, P., Treude, C., & Matsumoto, K. (2024). Generative AI for Pull Request Descriptions: Adoption, Impact, and Developer Interventions. ACM.

Zewe, A. (2023, November 9). Explained: Generative AI. Retrieved from mit news: https://news.mit.edu/2023/explained-generative-ai-1109