While hiring systems select for the past, ObjectivEye is building AI for the labour market of tomorrow

The labour market is changing faster than ever. Roles are less stable, skills shift continuously, and new competencies emerge more quickly than organisations can keep up with. Towards 2030, a large share of the skills currently in demand is expected to have changed. Yet hiring systems continue to select for the past, for CVs, job titles and historical patterns, while decisions increasingly need to be made on the basis of future potential.

That tension reveals a structural problem. And it is precisely the problem that Noura El Ouajdi, CEO of ObjectivEye, wants to solve.

A system that loses sight of talent

In practice, this tension has a concrete consequence: a large share of available talent simply remains out of view. Not because it is absent, but because the system cannot recognise it.

Current hiring processes are not designed to handle the scale and complexity of the modern labour market. They rely on proxies, CVs, educational background, familiar names, while recruiters are constrained by time and capacity. Combined with bias in both human assessment and automated systems, a structural visibility problem emerges.

The result is that talent is not selected on the basis of what someone can do, but on the basis of what the system can process.

"As long as hiring remains based on the interpretation of CVs and subjective judgements, we will keep scaling bias and inefficiency," says El Ouajdi. "We are not making the wrong decisions by accident, we have built a system that produces them with certainty."

The economic impact of decision-making without a foundation

This lack of structure and visibility translates directly into economic damage.

Traditional recruitment procedures take an average of 42 days. A bad hire costs an organisation an average of €29,000. And in Europe, 40% of newly hired employees leave their position within six months of joining.

These costs are not isolated incidents. They are the consequence of decision-making without a structural foundation. Organisations repeatedly make decisions based on fragmentary information and personal interpretation, leading to inconsistent outcomes, high turnover and a structural loss of talent.

"With ObjectivEye, organisations can accelerate their recruitment processes by up to 30% and significantly improve the match between candidate and role," says El Ouajdi. But speed is the least interesting part of the story. What matters is the quality of the decision itself.

The compliance pressure that accelerates everything

Beyond economic impact, pressure from regulation is also growing, and that is accelerating the urgency.

Recruitment falls under the high-risk AI application category within the EU AI Act. That means organisations must be able to explain how decisions are reached, which factors played a role, and how bias has been mitigated. Not as a best practice, but as a legal obligation.

Many existing systems are not set up for this. Decisions are often not transparent, audits are not possible, and bias is embedded in the processes. This creates a new type of risk: not only bad hires, but also insufficient control over how those decisions were made.

"We start with compliance by design," says El Ouajdi. "Not as an afterthought, but as the starting point of the entire architecture. In a context where recruitment is a high-risk application, explainability is not a differentiator, it is a prerequisite."

Skills as new infrastructure

The core of ObjectivEye is a redesign of how hiring decisions are constructed.

Instead of working with static job profiles and CV matching, the platform approaches recruitment through skills. Candidates and roles are described within a layered skills architecture, so that assessment is based on what someone can actually do, not on what title they held in the past.

This approach is directly connected to a labour market in which roles are constantly changing. "Skills architecture needs to be approached bottom-up," says El Ouajdi. "Flexibility is essential. A rigid framework imposed from the top down does not work, organisations are too diverse, roles change too quickly."

In practice, ObjectivEye uses each client's unique organisational metadata as its starting point. TNO is also developing a new national competency language, in collaboration with UWV, SBB, CBS and the Ministry of Social Affairs, which will be integrated into the platform. Standardisation and practical applicability are being built and tested simultaneously in realistic contexts.

Explainable AI: not a feature, but a requirement

Within this system, AI plays a central but controllable role.

Every decision is made transparent: which competencies were taken into account, where the differences between candidates lie, and what information is still missing. Both AI and HR staff assess candidate profiles. Low scores are accompanied by human feedback. The entire recruitment journey, from initial application to appointment, remains traceable.

"Transparency is fundamental to the product," says El Ouajdi. "Every step in the process must be explainable, not only to the candidate, but also to the organisation itself and to external supervisors."

Many existing HR tech solutions cannot offer this. Enterprise organisations that currently deploy AI for recruitment without an explainability function are running compliance risks that will only grow as the AI Act is increasingly enforced. ObjectivEye builds that in from the outset, as both a structural advantage and a necessity.

Governance: the human in the loop is not a side note

There is a misconception in the HR tech world: that AI replaces humans in the recruitment process. ObjectivEye does the opposite.

The platform explicitly integrates a governance layer in which human assessment, monitoring and correction are part of the decision-making system. In an environment where decisions must be explainable and auditable, that human-in-the-loop is not a nice-to-have, it is a prerequisite for responsible use.

"The governance layer is essential for complex organisations," explains El Ouajdi. "Public organisations have governance teams, maturity roadmaps, multiple stakeholders. You need not only data expertise, but also a structure to validate matches, make corrections and account for decisions."

Phased growth

ObjectivEye is being brought to market in phases through pilot projects. Technology, decision-making and governance are validated simultaneously, in complex environments with high requirements.

The first clients are immediately significant players: De Nederlandsche Bank, the Ministry of Social Affairs, Medisch Spectrum Twente and TNO itself. The Dutch Tax Authority (Belastingdienst) and Rijkswaterstaat are also considering joining.

These are not simple test cases. They are large, heavily regulated organisations with diverse job profiles and substantial recruitment needs. The choice for this type of environment is deliberate. Complex and regulated organisations represent the logical first step, precisely because the pressure on decision-making, compliance and scalability is greatest here. If the system works in these settings, it works everywhere.

The problem is not data. It is structure.

The current hiring system loses sight of talent, causes economic damage, fails to meet regulatory requirements, and does not align with a labour market that is changing faster than ever. Four dimensions, one structural failure.

ObjectivEye addresses all four, by redesigning hiring as a structured, explainable and scalable decision-making system. No longer: who seems suitable? But rather: which competencies underpin this decision, and how can we demonstrate that?

"We are not replacing recruiters," says Noura El Ouajdi. "We are giving them a system with which they can make decisions that hold up, for the candidate, for the organisation, and for the regulator."

ObjectivEye is a spin-off of TNO, launched in March 2025. The company is developing an AI-driven recruitment platform aimed at enterprise organisations, with explainable AI, skills-based hiring and compliance with the EU AI Act as its core pillars.