Mahala.ai targets Gulf AI talent bottleneck

Mahala. ai is expanding access to senior data and artificial intelligence specialists for organisations across the UAE and wider Gulf, targeting a skills shortage that has become a major constraint on companies trying to turn AI investment into measurable business results.

The Netherlands-headquartered company is offering enterprises and government organisations remote access to experienced data engineers, data scientists, AI and machine-learning engineers, and MLOps specialists. Its model is designed particularly for organisations that need specialised technical capabilities without lengthy recruitment, relocation or visa processes.

Mahala. ai says specialists can work within a client’s own virtual desktop infrastructure, keeping corporate data inside the organisation’s security perimeter. Its talent pool is aligned with Gulf working hours and includes Arabic- and English-speaking professionals, an approach aimed at regulated industries where data residency, intellectual property protection and access controls can complicate conventional outsourcing.

The company promises an initial response within 24 hours and a shortlist within 72 hours after receiving a detailed brief. Candidates are evaluated under a 100-point vetting system covering technical capability, demonstrated impact, consulting aptitude and professional development. A minimum score of 75 is required, with Mahala. ai saying roughly one in seven applicants qualifies for its talent bench.

The proposition arrives as UAE demand for AI expertise accelerates. The proportion of job advertisements requiring AI-related skills climbed from 1 per cent in 2021 to 3.2 per cent in 2025. About 12,200 UAE job advertisements required AI capabilities in 2025, up by roughly 2,700 from the previous year. The increase placed the country among the faster-growing AI talent markets internationally.

Demand is also spreading beyond technology companies. Financial services, professional services, manufacturing, energy and other industries are embedding AI capabilities in operations, while technology, media and telecommunications have the highest concentration of AI-related vacancies. Scarcity is feeding into compensation: jobs requiring AI skills command significant wage premiums in several UAE sectors, particularly financial services and technology.

Yet hiring more specialists addresses only part of the problem. Companies worldwide are discovering that ambitious AI programmes can fail when data, governance and production infrastructure are not ready to support them. Through 2026, an estimated 60 per cent of AI projects unsupported by AI-ready data are expected to be abandoned. A survey of 1,203 data-management leaders also found 63 per cent of organisations either lacked, or were uncertain whether they possessed, suitable data-management practices for AI.

That distinction is important. Successful AI deployment increasingly requires engineers who can build reliable data pipelines, manage model serving and observability, integrate systems with existing applications and maintain governance after models enter production. Mahala. ai has consequently divided its offering across data engineering, data science and analytics, AI/ML engineering, and MLOps and AI infrastructure rather than concentrating solely on generative-AI development.

Evidence from enterprise generative-AI adoption has reinforced concerns about the gap between experimentation and financial returns. MIT’s 2025 GenAI Divide study examined 300 publicly disclosed AI deployments, conducted 150 interviews and surveyed 350 employees. It found that only about 5 per cent of integrated enterprise AI pilots were generating rapid measurable financial impact, leaving the overwhelming majority without demonstrated profit-and-loss gains at the stage studied.

The finding does not mean that 95 per cent of all AI technology fails. The study instead highlighted difficulties integrating systems into workflows, retaining organisational context and adapting tools to operational needs. Enterprise AI programmes that crossed the gap tended to focus on specific business problems and stronger integration rather than deploying generic tools without changes to processes and data architecture.

For Mahala. ai, that implementation gap creates an opening beyond conventional recruitment. Its specialists can be deployed as project executors, strategic advisers or end-to-end technical owners, while the company handles cross-border contracting, intellectual-property provisions and invoicing. Co-founder and chief executive Amir Grabic and co-founder and executive director Jos Preusting oversee a model that emphasises small curated shortlists rather than large volumes of CVs.