Defence, NewSpace, medtech: you are all hiring the same engineers

A surgical robotics start-up, a defence contractor and a NewSpace scale-up can all be chasing the exact same candidate, at the exact same time, without ever crossing paths at a trade show.

Embedded systems engineer. RF specialist. Signal processing expert. Electronics or mechatronics engineer.

These 4 job titles keep showing up in briefs that have nothing to do with one another. And whenever we open a search for one of these 3 sectors, the shortlist of companies where the candidates we're tracking currently work includes the other two.

The problem isn't that the pool is small. It's that it's small and shared.

Key takeaways

• Defence, space and medtech are hiring for the same skills: signal processing, radio frequency, real-time embedded systems, electronics under tight size and power constraints.
• The pool isn't growing as fast as demand. Specialised courses remain rare, and signal processing stays a niche within engineering degrees that are mostly generalist.
• Demographics make the imbalance worse. 48% of jobs in the defence industry are held by people aged 43 and over, and of the 100,000 hires targeted by 2030, 70,000 will go towards replacing retirements.
• Some skills can't be replaced: engineers who master radar systems designed in the 1980s and 1990s, still in service today, are retiring with no trained successors.
• In 9 cases out of 10, these profiles are already employed at a large group, don't browse job boards, and don't reply to ads.
• The practical consequence: direct approach isn't one option among others, it's the only one that works, and it's won on the project rather than the salary.

Why are all 3 sectors after the same engineer?

Because they're solving the same physical problem.

A radar, an Earth-observation satellite and a surgical guidance system are all fundamentally doing the same thing: capturing a weak signal, processing it in real time, and fitting within a constrained volume, constrained power draw and a reliability bar that tolerates no failure.

The skills that answer this brief are identical from one sector to the next:

• Signal processing and embedded algorithms
• Radio frequency and microwave engineering, antenna design, electromagnetic compatibility
• Embedded electronics under size, weight and power constraints
• Real-time systems and functional safety, with the certification cultures that come with them

An engineer who designed a radar receiver chain at a defence equipment maker is directly employable on a satellite payload, and vice versa. The gap is in the application domain, not the technical foundation.

This is precisely what dual-use technology means: the same building blocks serve civilian and military purposes. On the hiring side, duality has a less-discussed effect: it puts companies that don't see themselves as competitors into direct competition with each other.

Why isn't the pool growing as fast as demand?

Two reasons combine, and the second is the most underestimated.

Training produces few specialists

French engineering schools train generalists first and foremost. Signal processing, RF and microwave engineering remain final-year options or lab specialities, carried by a small number of institutions.

The result: each graduating class feeds the market a trickle at a time, while demand has changed scale entirely. France's 2024-2030 Military Programming Law, updated to €436 billion by Parliament in July 2026, and the growth of the private space sector, have opened roles far faster than course places have opened up.

The age pyramid is tipping over

This is the factor people talk about least, and it weighs the most.

The national plan for employment in the defence industry, launched by France's Ministry of Labour on 15 September 2026, spells out the stakes bluntly: 100,000 hires to make by 2030, of which 70,000 will be needed to replace retirements. The same document notes that 48% of jobs in the defence industry are held by people aged 43 and over.

In other words, most of the hiring effort of the coming years won't be about growth. It will be about not shrinking.

What happens when a skill retires?

For some profiles, replacement isn't even really on the table.

Take radar systems designed in the 1980s and 1990s, still in service today. The engineers who master them are now between 50 and 60 years old. Their expertise was built over decades, on specific architectures, often with no documentation a newcomer could actually use.

These skills can't be hired for, because they're no longer taught. Nobody teaches a 1987 radar architecture today.

The consequence is twofold for the companies concerned. They struggle to maintain existing systems, and they struggle to hire to evolve them, since modernisation first requires understanding what's already in place.

This is the only segment of this market where hiring isn't enough on its own. Knowledge transfer has to be organised before the departure, not after.

Where are these engineers, and why aren't they moving?

They're not impossible to find. They're elsewhere.

In nearly every search we run on these skills, the profiles we identify are already employed, at Thales, Safran, Airbus, Dassault, MBDA, Naval Group, ArianeGroup, ONERA (France's aerospace research agency), the CEA (France's atomic energy and alternative energies commission), electronics equipment makers, or private space pure players.

They're not applying. They're not browsing job boards. Some haven't opened LinkedIn in months.

This isn't a lack of interest in the market. It's a simple mechanism: a large group offers stability, long programmes and a level of security few start-ups can match on paper. Even once the project has stopped exciting them, the trade-off stays stacked against moving, as long as nobody comes and reframes it for them.

And this is where the asymmetry becomes visible. France's private NewSpace sector counts around 150 private companies for just over 2,000 direct jobs, according to CNES, France's national space agency. A 40-person scale-up ends up competing for a signal engineer against a group that employs tens of thousands, with an employer brand on an entirely different scale.

It can win. Just not on the same arguments.

What does this change in practice for your hiring?

Four practical consequences, which we see on roughly every mandate of this kind.

1. The job ad will produce nothing. Posting a role for these skills means waiting for the 10% of the pool that's actively moving, which is exactly the slice every one of your competitors will see too.
2. The timeline runs in months, not weeks. An engineer in post on a long programme doesn't decide in 15 days, and their notice period adds to that. On a security-cleared role, the administrative investigation stretches the calendar further still, as we detailed in our article on hiring a defence engineer and security clearance.
3. The project matters more than the salary. On these profiles, the argument that actually moves people is almost always technical: a subject they can't work on where they are, a level of responsibility out of reach within a large programme, a product that will actually go into service rather than stay in study phase.
4. Widen the sector scope of your search. If you're hiring for defence, look at space, scientific instrumentation and medtech too. The duality of these skills works in your favour as well.

This last point is the most actionable, and the least often applied. The briefs we receive are still almost always framed by sector of origin, while the skill itself recognises no such boundary.

The same mechanism holds outside engineering, too. On sales roles, insisting on sector experience creates exactly the same bottleneck, as we've covered in our article on hiring a business developer in defence.

In summary

Defence, space and medtech aren't competing in their markets. They're competing in their hiring, without always realising it.

The shared talent pool is narrow, it renews slowly, and every year it loses skills that training no longer reproduces. Faced with that, the one lever a company fully controls is its ability to go after people who aren't looking, and to offer them something other than a role equivalent to the one they already hold.

Hiring for these skills? Bluecoders covers all 3 worlds: the Defence, Industry & Hardware practice for defence, aerospace and space, and the practice dedicated to medtech. Let's talk about your need.

FAQ

Why are signal processing engineers so hard to hire?

Because supply and demand are moving at opposite speeds. Signal processing remains a niche speciality within engineering courses that are mostly generalist, so the flow of new graduates is low and stable. On the other side, defence, private space and medtech have all increased their hiring needs over the same period. A demographic effect adds to this: in the defence industry, 70,000 of the 100,000 hires targeted by 2030 will go towards replacing retirements.

Do defence and space really hire for the same profiles?

On the technical foundation, yes. A radar and a satellite payload pose the same problem: capturing a weak signal, processing it in real time, fitting within constrained volume and power draw. An engineer from one is employable on the other after ramping up on the application domain. It's one of the concrete forms dual-use technology takes.

How do you hire an engineer who's already employed at a large group?

Through direct approach, and on a technical argument rather than a financial one. These profiles don't browse job boards. What moves them is almost always a subject they can't work on where they are: a level of responsibility out of reach within a large programme, a product that will genuinely go into service, or a design freedom an organisation of several thousand people can't offer.

Can a start-up hire against Thales, Safran or Airbus?

Yes, but not on the same ground. A start-up won't win on stability, nor on brand recognition, nor, most of the time, on package. It wins on the scope entrusted to the hire, speed of execution, and visibility of the outcome. An engineer running a sub-function within a 10-year programme can move for a role where they own the full architecture of a product.

What should you do about skills on legacy systems that are retiring?

Treat it as a knowledge-transfer issue, not a hiring one. 1980s and 1990s radar architectures are no longer taught, so the market won't produce a replacement. Companies in this position organise the handover ahead of the departure: pairing, documenting design decisions, and hiring a more junior profile trained internally by the person who's leaving.

Should you widen your search to other sectors to find these profiles?

It's one of the most effective, and least used, levers available. The duality of these skills cuts both ways: if it puts defence, space and medtech in competition, it also opens each of them up to the other two's talent pool. A brief framed around the skill rather than the sector of origin mechanically widens the target, without lowering the technical bar.

How long does it take to hire an RF or signal processing engineer?

Longer than most engineering roles, because 3 delays stack up: the time to convince someone who wasn't looking, the notice period of someone already employed, and, on a role exposed to classified information, the administrative investigation ahead of security clearance, whose timeline is in neither the employer's nor the candidate's control.