AI-assisted software development is reshaping South Africa’s technology workforce, with growing concern that entry-level developers risk losing the critical opportunities they need to build foundational engineering skills — a shift that could damage the country’s tech sector if left unaddressed.
The integration of AI into software development is not merely another industry trend but a fundamental restructuring of how technology is built, who builds it, and how engineering talent is cultivated, according to Lisa Jasper, head of talent acquisition at Dariel.
“For South Africa, a nation that has positioned technology as a critical engine for economic growth and social mobility, understanding and responding to this shift is essential,” she said. “Unlike previous disruptions, this transformation demands a strategic, multi-stakeholder response.”
The pace of change is striking. According to GitHub’s 2023 survey, 92% of developers in the United States now use AI coding tools either at work or in personal projects, with 70% reporting significant advantages in accelerating learning and boosting productivity. McKinsey’s 2023 research suggests software engineering is among the functions most exposed to automation, with potential productivity gains of 20% to 45% in coding tasks. While the figures originate from global studies, the implications ripple directly into markets such as South Africa, where technology services are increasingly integrated into global delivery models.
The stakes locally are high. South Africa’s ICT sector contributes roughly 8% to national GDP and employs hundreds of thousands of workers directly and indirectly. For many — particularly graduates from historically disadvantaged backgrounds — the technology industry has been a critical pathway into the middle class. Youth unemployment remains above 45%, and the tech sector has been one of the few consistent sources of quality employment for young graduates. South Africa’s National Digital and Future Skills Strategy identifies digital skills as a national priority, but Jasper argues that the rapid evolution of AI threatens to outpace existing educational and training frameworks.
South Africa’s technology sector has weathered global disruptions before — the Y2K remediation boom, the dot-com collapse and the 2008 financial crisis all affected hiring and investment flows. But Jasper argues that this disruption is different. “When global markets recovered in the past, local demand returned, and software engineers continued to follow a familiar growth path: junior roles, skills accumulation, and long-term career stability,” she said. “The AI transformation does not follow this pattern. Rather than temporarily suppressing demand, it is fundamentally altering the composition of software teams and the nature of work itself.”
At the heart of the issue is a troubling trend: the disappearance of traditional junior developer tasks. “We are seeing an erosion of entry-level development work,” Jasper said. “AI-assisted tools — including GitHub Copilot, Amazon CodeWhisperer and emerging open-source alternatives — are increasingly capable of generating boilerplate code, automating routine bug fixes, completing basic feature development, and writing documentation and test cases.”
The problem is that these tasks have traditionally served as the training ground for junior developers — the proving ground where foundational skills are built. A 2024 Stack Overflow Developer Survey found that 76% of developers are now using or planning to use AI tools, with code generation and debugging assistance being the most common applications. For emerging markets such as South Africa, where entry-level roles have served as on-ramps to professional careers, the shift carries significant implications. “The concern is not that AI will eliminate engineers,” Jasper said. “But that future engineers may advance without developing deep problem-solving and diagnostic skills.”
Not everyone views the transition with alarm. Many industry leaders argue that AI will augment rather than replace developers — Microsoft CEO Satya Nadella has consistently positioned AI tools as productivity enhancers rather than replacements. From that perspective, developers who embrace AI may become significantly more productive, enabling them to tackle more complex and creative challenges. “Developers who master AI-assisted workflows may become more competitive in global markets, potentially attracting more international investment and remote work opportunities,” Jasper acknowledged.
But labour economists and workforce development specialists urge caution. The World Economic Forum’s Future of Jobs Report 2023 projects that 44% of workers’ skills will be disrupted within five years, warning that significant portions of the workforce risk displacement without proactive reskilling. In South Africa, where skills development infrastructure is already strained, Jasper sees particular cause for concern. “The risk of a ‘missing middle’ — a generation of engineers who never develop foundational competencies — is a genuine concern.”
From a social development standpoint, the AI transition raises questions of fairness. South Africa’s technology sector has been a rare bright spot in an economy marked by persistent inequality. If AI-driven efficiency gains accrue primarily to capital and senior talent while entry-level opportunities contract, the sector’s role as an engine of social mobility could be undermined.
Jasper argues that the answer lies not in resisting AI but in reshaping how the industry develops talent alongside it. She outlines several priorities. The first is to redesign early-career pathways: “Rather than eliminating junior roles, forward-thinking organisations should redesign them. Entry-level positions should evolve into ‘AI-enabled apprenticeships’ that emphasise system design and architecture thinking, code review and quality assurance, security awareness and governance, and collaboration with AI tools as a core competency.”
The second is to invest in structured mentorship. “In an AI-augmented environment, the tacit knowledge transfer that occurs through mentorship becomes more critical, not less,” she said. The third is to prioritize responsible AI integration. “Adopting AI tools without considering their impact on workforce development is short-sighted. Organisations need to establish clear policies that balance productivity gains with talent development objectives.”
The fourth is updating national skills frameworks. “From a government perspective, South Africa’s existing digital skills strategies must be reviewed and updated to reflect the realities of AI-augmented development,” Jasper said. “This includes identifying emerging skill requirements and ensuring that public training programmes address them. Government should establish mechanisms to track the impact of AI adoption on technology employment, enabling evidence-based policy responses.”
Ultimately, Jasper frames the challenge as one of alignment. “The AI-driven transformation of software engineering presents both risks and opportunities for South Africa,” she said. “The risk is that efficiency gains come at the expense of long-term capability development, hollowing out the talent pipeline and undermining the sector’s role in economic inclusion. The opportunity is to lead in defining how human expertise and artificial intelligence can work together productively and responsibly.”
She closed with a note of cautious optimism. “AI has clarified where human value is most critical. By aligning industry practices, educational approaches, and policy frameworks, South Africa can continue to cultivate a technology workforce that is not only globally competitive but also deeply skilled, ethically grounded and resilient in the face of continued change. The path forward requires collaboration, foresight and a commitment to ensuring that technological progress serves broad-based prosperity.”
Dariel Warns AI-Assisted Coding Could Hollow Out South Africa's Junior Developer Pipeline
This piece is based on an article originally featured on Intelligent CIO Africa: "South Africa risks losing a generation of software engineering talent in the AI era" by Bill Tanner.
