Enabling and not replacing: How AI is freeing up devs to do more impactful work

In this article, we explore how South African developers are using AI tools in their development process and what this says about AI: Is it replacing them or enabling them to do more?

Exclusive Dev Nation sneak peek: More than 40% of South African developers still aren’t using AI tools

In 2023, AI adoption in South Africa was still pretty slow, but this picture has changed significantly over the past twelve months.

This rapid increase in AI adoption can be seen in less experienced programmers becoming more productive when working with a tool like Co-pilot, as well as tech companies reducing engineering hours spent on manual work. For example, Duolingo reduced their median code review time by 68% by using AI tools as part of their development process.

From the data, we can also see the rise in AI adoption: More than half of developers have worked with an AI API, significantly up from just over a third in 2023.

"Have you used any AI APIs in the last year?"

While the majority of developers have also worked with a code-writing assistant in either their roles or personal projects,, over 40% of software developers still have no experience with these tools.

"Are you or your company using co-pilot or other AI code-writing assistants?"

This made us ask: What productivity gains are these companies and developers missing out on? And, does adopting AI enable developers or slowly replace them?

Let’s take a closer at how South African developers are using these tools to unpack each of these.

Developers are using AI to find insights and automate repetitive work

"How do you use AI tools as part of your role or personal projects?"

The most common way developers use AI tools is for data analysis and insight generation, followed by automating repetitive tasks and improving development processes.

In both of these cases, AI is taking over time-consuming tasks that previously required valuable developer time. If you haven’t adopted these tools yet, here are some of the tasks AI is automating so that developers can focus their attention on more complex technical problems:

• Automating testing of software applications which minimises the number of manual tests developers need to run.
• Assist developers with code debugging so that less time is spent identifying and fixing bugs.
• Optimising code performance by highlighting potential bottlenecks and improvements.
• Auto-generating documentation for developers so that they’re spending less time creating necessary documentation.

The type of developer work that AI is automating encompasses only a small fraction of the software development lifecycle. As a result, developers aren’t being replaced. Instead, they’re being enabled to do more of the work that can’t be automated.

While AI isn’t replacing engineers or threatening job security, it’s changing what roles look like. Developers who pick these skills up now are ahead of the curve to become the most valuable engineers.