Emerging tech roles are rarely brand-new jobs. They are combinations of skills from existing disciplines that companies need before the market has agreed on a title. If you hire or job-hunt by title alone, you will miss the people and the roles that matter most.

For engineers, this means the next step in your career may sit under a familiar label like backend or platform engineer. For hiring managers and CTOs, it means a vacancy for an "AI engineer" often describes three or four jobs at once. Both sides run into the same problem: the title says one thing, the work says another.

They are recurring skill bundles that cut across disciplines, such as ML engineering plus DevOps, or cloud plus reliability plus cost control. The title comes later. The work comes first.

Andela analysed 47,101 technical job postings from Fortune 500 companies and mapped 2,026 distinct skills. They found 23 recurring bundles that did not fit standard titles: eight new roles, 14 hybrids, and one that was left out of the final analysis.

Some examples: the MLOps Pipeline Engineer, the LLM Application Engineer, the FinOps Reliability Engineer and the Docs-as-Code Engineer. Others are the Product Front-End Engineer, the Lakehouse Analytics Engineer, and the DevSecOps and SecOps Observability Engineers.

Note the limits. This is US enterprise data, and a skill count is not a headcount. It shows where the work is moving. It does not prove how many people already do it.

Because companies advertise a familiar label while the work requires a mix. An "AI engineer" vacancy can ask for software engineering, data pipelines, evaluation, cloud operations and security in one go.

Of roughly 1,832 postings labelled "AI engineer" or "ML engineer", 53% required skills from at least two established roles. Andela's analysis

The same research estimates that 6,758 postings contained the bundle of an LLM Application Engineer without using that title. That is a role someone is hiring for, in an advert that never names it.

Corey Hymel of Andela put the risk well: any gap between the role a company thinks it is hiring for and the role the work requires becomes a structural liability.

In practice we see this at intake. The hiring manager says "we need an ML engineer". Twenty minutes later it turns out they need someone who gets models into production and keeps the cloud bill under control. That is another person.

Each one has a clear core discipline plus a few adjacent skills. The mix tells you where to search.

The MLOps Pipeline Engineer connects model development to production. The skill mix is roughly 46% ML engineering and 23% DevOps, with data engineering at 15%. The LLM Application Engineer builds products on top of foundation-model APIs, retrieval and agents. That is about 48% AI engineering and 34% ML engineering. It is not the same as training frontier models.

The FinOps Reliability Engineer treats cloud cost as a reliability problem. Think 36% DevOps, 27% site reliability engineering and 18% cloud engineering. The Docs-as-Code Engineer turns documentation into structured, version-controlled material that developers and AI systems can both use.

Look at the percentages again. Every role has one dominant core. The adjacent skills are added on top. That is why a strong infrastructure background often travels well, for example in a role like senior infrastructure engineer.

Not toward pure generalists. The strongest candidates for these hybrid roles usually keep a deep core specialty and add neighboring skills. Hybrid does not mean generic.

You hear it a lot: everyone is becoming a generalist. The research does not support that. Companies want depth in one area plus enough range to work across the seams between teams.

Hymel also argues that the cost of code is heading toward zero. If that holds, other things weigh more: system design, prioritization, communication and deployment experience. Those are the things that separate someone who writes code from someone who owns a result in production.

One caveat. This is one study and one interview-style discussion, so treat the direction as solid and the exact weights as indicative.

For engineers, the practical move is simple. Keep your core. Add the skill next to it that your current team keeps outsourcing to someone else.

Dutch demand is moving, but slowly and inside existing stacks. AI skills are absorbed into cloud, data and software work rather than replacing it.

According to Dutch job-posting data, vacancies requesting AI-related skills rose from 1.5% of postings in 2022 to 2.1% in 2025. That is growth, but from a small base. In the 90 days to mid-October 2026 there were 35,898 technology vacancies, led by Project/Program Manager and Backend Engineer. Python, Azure and SQL were the top technical skills.

So the market is still anchored in familiar roles. If you run a Dutch scale-up, the candidate you need may currently be titled backend, platform, data or DevOps engineer. Someone with Azure depth, for instance an Azure engineer, may be closer to an MLOps or FinOps profile than any "AI" title suggests.

EU rules on privacy, security and AI governance add to this. Roles that combine delivery with documentation, observability and risk control get more weight. Salary data for these hybrid titles is thin. Be careful with any single number you see for emerging tech roles salary.

Describe the outcome and the task bundle, not the tool list. Name the one core discipline you must have and the two or three adjacent skills you would like.

The common mistakes are easy to spot:

• Recruiting by title instead of by what the person will actually do.
• Listing every framework, which hides the real need, such as reliable model deployment or controlled inference costs.
• Mixing senior ML research, platform engineering, security and product management into one wish list.
• Screening for syntax when production judgment is what matters.

An impossible wish list shrinks the pool and pushes pay expectations up. It also makes interviews inconsistent, because nobody agrees what "good" looks like.

Here is how we handle it in an intake. We ask what the person has to ship in the first six months, who they depend on, and what breaks if they do it badly. From that we write the skill bundle, then decide where to search. It often leads us to engineers who never had "AI" in their title, and sometimes to a profile like an artificial intelligence engineer where the real need is a mix of production and data skills.

For your resume, the same logic applies. Describe what you shipped and operated, not only the tools you touched. That matters more for emerging tech roles than a fashionable title.

Emerging tech roles show up as skill combinations long before they get a name. The sensible move is to find the durable core, map the adjacent skills the business really needs, and hire for that bundle. Engineers can do the same with their own profile.

Are you trying to fill a role that does not fit any standard title? We start with the task bundle in intake and work from there. Happy to think it through with you.

1. Tech Hiring Trends in Netherlands 2026 - Get U Hired
2. Andela skills index exposes AI talent misdirection
3. LLMs Topic - Yomu
4. career - Stack Overflow
5. Tales from the 2026 Developer Survey results - Stack Overflow
6. Companies hiring ai now in Europe — 853 Companies | SignalsAPI

Written by our AI, read by a flesh-and-blood recruiter.