Demand for data engineers and AI specialists has grown explosively. But precisely in a young, fast-moving market it is hard to separate the wheat from the chaff. Everyone puts "AI" on their profile. What do you look for to really hire the right specialist?
In this article you will learn what to look for when hiring a data engineer or AI specialist: which skills and experience matter, how to keep realistic expectations, and how to arrange the Dutch DBA Act. That way you choose on substance, not on buzzwords.
This blog is for clients: IT managers and decision-makers who want to hire data or AI expertise and choose the right one.
This is a young and fast-changing field, so this blog gives guidelines, not absolute truths. The market moves quickly; check the current state for important decisions.
What is the difference between a data engineer and an AI specialist?
A data engineer builds and manages the infrastructure with which data is collected, stored and made available. An AI specialist uses that data to build, train or implement AI solutions. The roles overlap, but the emphasis differs: data engineering is foundation, AI is application.
For your hiring, this distinction is important, because you often first need a solid data foundation before AI applications make sense. So determine sharply which problem you want to solve. Do you seek someone to get your data in order, or someone to build a concrete AI application? That determines which profile you need.
Which skills really matter?
That depends on your need, but there is a core. For data engineering that is building data pipelines, databases and cloud data work. For AI it is about experience with actually building, training or implementing models and solutions, not just theoretical knowledge.
Watch especially demonstrable, practical experience instead of popular terms on a CV. In this field, many people claim they do "something with AI", while real implementation experience is scarcer. Probe on concrete projects: what has someone really built, with what result, and what problems did they run into?
How do you separate a real specialist from a buzzword profile?
By probing on concrete results. Have the specialist describe a project they did themselves: what was the problem, what did they build, which choices did they make and what did it deliver? Those who really did it talk about data, trade-offs and obstacles, not just about the technology itself.
Be alert to people who mainly talk in terms and hypes without concrete examples. A good intake exposes this difference. Also ask about what did not work, because being able to honestly name what was difficult is often a sign of real experience. Want to know how to conduct such a conversation? Read our blog on assessing a freelancer in an intake.
Which expectations are realistic?
Be realistic about what AI can deliver in your situation and on what timeline. The market is full of promises, but real value often arises step by step, starting with a solid data foundation. A good specialist will temper and adjust your expectations rather than promise everything.
See it as a warning signal when someone promises big results without first understanding your data, your situation and your goals. Realism is a sign of quality here. A specialist who is honest about what is and is not feasible is more valuable than one who promises you riches. Reliability weighs extra heavily in this young field.
What does a data engineer or AI specialist cost?
These profiles are among the scarcest and thereby often best-paid in the market, especially with demonstrable implementation experience. The scarcity is driven by the high demand and the limited supply of people who have really put AI into practice. So count on firm rates for proven expertise.
Weigh that rate against the value and the risk: an expensive specialist who does it well is more economical than a cheap one who lets your project fail. Want a picture of the rates for scarce IT profiles? See our guide on what hiring an IT professional costs.
How do you arrange the Dutch DBA Act for data and AI hiring?
As with other IT hiring, the Dutch DBA Act applies: if you hire a self-employed person, the relationship must demonstrably be that of an assignment, not of employment. Data and AI projects are often long-running and intensive, so this requires attention to definition and independence.
Steer on a defined result, avoid embedding as a permanent team member, and make sure the practice matches the contract. For long-running projects, the intermediary construction can cover the risk while the specialist stays independent. Want to know how the assessment works? Read our guide to the 9 assessment factors of the Dutch DBA Act.
Frequently asked questions about data and AI hiring
Do I need a data engineer first or an AI specialist first?
Often a data engineer first. AI applications need a solid data foundation to be meaningful. If your data is not in order, an AI specialist delivers little. Determine sharply which problem you want to solve; that determines whether you need the foundation or the application first.
How do I recognise a real AI specialist?
By demonstrable, practical implementation experience, not by buzzwords. Probe on concrete projects someone built themselves, with what result and against which obstacles. Those with real experience talk about data and trade-offs; those who mainly talk in hypes without examples are a signal to be careful.
Are data and AI specialists more expensive than other IT professionals?
Usually yes. Due to the high demand and limited supply of people with real implementation experience, they are among the best-paid profiles. Count on firm rates for proven expertise. Weigh that against the value: a good specialist pays for itself, a mismatch is expensive.
What if the AI solution does not deliver what was promised?
You reduce that risk by setting realistic expectations in advance and working with a specialist who is honest about what is feasible. Start small and build step by step. A specialist who promises big results without knowing your situation is a risk; realism is a sign of quality here.
Can I hire a data or AI specialist as a freelancer?
Yes, provided you arrange the Dutch DBA Act well. Steer on a defined result, avoid embedding, and make sure the practice matches the contract. Because these projects are often long-running, the intermediary construction can cover the risk for longer engagements while the specialist stays independent.
Conclusion: choose on substance, not on buzzwords
Hiring a data engineer or AI specialist in this young market comes down to one thing: separating the wheat from the chaff. Watch demonstrable, practical experience, probe on concrete results, keep realistic expectations, and arrange the Dutch DBA Act well. That way you choose on substance instead of hype.
For whom is this most relevant? For clients hiring data or AI expertise in a market full of promises. For whom less? For those with a clear, defined and proven profile in mind who outsource the selection.
My advice: determine sharply which problem you want to solve, test for real experience, and be wary of big promises without substantiation. In a young field, proven experience is worth more than an impressive-sounding profile.
Help finding the right specialist?
Want to spar about which data or AI profile fits your need, or hand over the pre-selection? Plan a no-obligation call with me. I help you find the right, proven specialist.
Note: this is a fast-changing field. This article is a snapshot of early 2026. Regulations around the Dutch DBA Act may change; consult rijksoverheid.nl or belastingdienst.nl.
