Programming finished 23rd out of 26 core competencies in the World Economic Forum's Future of Jobs Report 2025, and demand for leadership and social influence surged 22 percentage points in two years. Those two data points, cited by TechCabal, are the clearest summary of a structural hiring crisis now playing out at engineering teams across the continent: companies are letting go of technically brilliant engineers who cannot translate their output into business value.

The argument, made by Mohamed Moniem — General Manager for North Africa at ALX Africa and former petroleum engineer who co-founded the digital transformation firm pyramidBITS — is that raw programming ability has been demoted to a baseline entry requirement. What actually determines retention and promotion is a different skill set entirely: strategic thinking, stakeholder communication, and the ability to direct automated systems toward the right problems rather than simply execute on syntax.

The economics of this misalignment are severe in North Africa specifically. A survey of leading Egyptian enterprises found that 78% of firms cannot find talent with adequate operational capabilities — creative thinking, problem-solving, and related human skills — not technical ones. In Morocco, the High Commission for Planning recorded 282,000 new jobs created between early 2024 and 2025, yet 37.7% of university graduates remained unemployed. The labour market is generating positions; young professionals are simply not equipped to fill them on the dimensions that now matter.

Generative AI is accelerating the crisis rather than creating it. As automated tools absorb initial code generation and routine builds, the execution phase of software development — the phase that traditional coding assessments measure — is being industrialised. What remains irreducibly human is precisely what standardised curricula do not teach: deciding which problems are worth solving, securing organisational buy-in, and leading teams through continuous technological change. An engineer who cannot do those things is, in Moniem's framing, a liability that requires a dedicated interpreter between their output and the rest of the business.

The lifespan of hard technical skills is collapsing alongside these shifts. Specific programming languages and frameworks can become obsolete faster than a new hire completes their probation period. That makes the capacity for self-directed, continuous learning the only durable technical asset — and it is a quality that no standardised coding test can reliably detect.

ALX Africa, where Moniem has built the Egypt business since 2022, is trying to operationalise a different model. The organisation uses a 10-20-70 learning framework: 10% traditional instruction, 20% peer interaction, and 70% practical application. Graduation is conditional on building functional systems and defending architectural choices before peers, forcing learners to practice precisely the communication and strategic reasoning skills that employers say they cannot find. The outcomes reported are notable: over 60% of ALX graduates secure employment within six months, rising to over 90% within a year, with hiring partners returning consistently to request more candidates from the programme. In three years, the Egypt operation alone has produced more than 29,000 graduates.

For engineering leaders and recruiters, Moniem's prescription is operational and immediate. Hiring interviews should require candidates to reason aloud about past projects — including architectural decisions that failed — rather than pass isolated coding challenges. Communication ability should be a binary gate, not a soft preference: a candidate who cannot explain their work to a non-technical stakeholder should be rejected on that basis alone, just as they would be for failing a technical screen. And the most predictive signal of five-year performance, he argues, is what a candidate has taught themselves in the past twelve months — an indicator of the self-directed learning habit that outlasts any specific skill.

Why it matters: The 78% talent-gap figure in Egypt and Morocco's 37.7% graduate unemployment rate are not soft complaints about soft skills — they represent a structural mismatch that is already costing African tech companies in turnover, lost velocity, and the overhead of maintaining human translators between engineering and the rest of the business. As AI compresses the value of pure coding further, any firm still screening primarily on technical assessments is optimising for the cheapest and fastest-depreciating input in its stack. Investors evaluating engineering teams, and founders building them, should be asking a harder question: not what languages the team writes, but whether they can explain why they wrote anything at all.