On 24 September 2025 the French economy ministry named the winner of an offshore wind tender in the Channel and, in the same release, recorded that a second tender off the Atlantic coast had received no offer at all. One park of 1,500 MW will be built more than 40 km off the Bay of Seine. Off the island of Oléron, nobody bid.
Those two sentences, printed side by side by the same ministry on the same day, are a better guide to the energy industry than any list of announced projects. They say that a project is not one thing. It is a sequence of states, and a job attached to a project in one state is not the job attached to it in another.
This year, Europe’s energy industry is building in three different states at once: offshore wind is awarded faster than it is built, electrolysers are committed at a small fraction of what was announced, and battery recycling is waiting for batteries that have not yet reached the end of their lives, so a candidate who can tell which state a project is in reads a job advertisement better than one who reads the announcement.
This is the third piece of the energy chapter of the CareerOn Industry Atlas. The first asked where the money is made; the second asked which positions the power system cannot replace. This one asks what is actually being built, and it holds itself to a simple rule. Every project is placed in one of five states: announced, awarded, final investment decision, under construction or operating. A project is placed in the highest state a dated document can prove, and never higher.
The piece reads eight documents: the French statistics service’s wind dashboard, the economy ministry’s tender release, the International Energy Agency’s hydrogen review and its outlook for electric vehicles, two reports of the European Court of Auditors, the annual barometer of the French hydrogen industry body and a decision of the French public debate commission. None of them counts jobs. We do not count them either.
Why the state matters more than the headline
An announcement is a statement of intent by a company. An award is a decision by a public authority that a named bidder will build a named project under named terms. A final investment decision is the moment money is committed and contracts for equipment are signed. Construction is when the site fills with people. Operation is when the asset produces, and the work changes again, from building to running and maintaining.
Each state calls for different people. Before an award, the work is studies, surveys, permitting and the preparation of a bid. Between award and investment decision, it is engineering, supply negotiations and financing. During construction it is site management, logistics, quality and safety. In operation it is performance, maintenance and compliance. A reader who knows only that a project exists does not know which of these a company is hiring for, or whether it is hiring at all.
That is why this piece refuses the most common shortcut in energy coverage, which is to add announced capacity together and call the total a boom. Announced capacity is a promise. The documents below show, family by family, how far the promises have travelled.
Offshore wind: more waiting than turning
At 31 December 2024, the French statistics service counted 1.5 GW of offshore wind connected to the grid and 3.4 GW of offshore projects under permitting review. The offshore fleet in operation was, in other words, smaller than the offshore fleet waiting for its permits. The same dashboard does not say which parks are under construction or when each will connect, and we do not infer it.
The ministry’s release adds a third state. Centre Manche 2 has been awarded; it sits near the earlier Centre Manche 1 project. An award is a firm public decision, but it is not a building site. Between the two lie design, surveys, supply contracts for turbines, foundations and cables, and the grid connection, each of which can take longer than the award itself did.
French offshore wind: more waiting than turning
| State | What the document prints | Date |
|---|---|---|
| Operating | 1.5 GW connected at sea | 31 December 2024 |
| In permitting | 3.4 GW of offshore projects under review | 31 December 2024 |
| Awarded | Centre Manche 2, 1,500 MW, more than 40 km offshore | 24 September 2025 |
| Tender with no bid | Oléron 1 (AO7) received no offer | 24 September 2025 |
Sources 1 Service des données et études statistiques (SDES) · 2 Ministère de l’Économie, des Finances et de la Souveraineté industrielle, énergétique et numérique
The Oléron tender is the most instructive row in the table, and also the easiest to misread. The release records that no offer was received at the end of the application period. It does not say why. It would be easy to supply a reason, cost of capital, turbine prices, the difficulty of the site, and each of those has been argued elsewhere. None of them is in the document, so none of them is in this piece. What the document proves is narrower and still useful: a public tender for an offshore park can close with no bidder, and a project map that shows every tendered zone as future capacity is overstating what is on its way.
For someone choosing a career, the shape matters more than the totals. France has more offshore wind in the pipeline than in the water. The work of the next years is therefore weighted towards the stages before operation: turning awards into projects, and permits into sites.
Hydrogen: the ambition shrank, the committed base still grows
No family of projects has been announced more loudly, or revised more visibly, than low-emission hydrogen. The International Energy Agency’s latest review puts potential low-emissions production from announced projects at 37 million tonnes per year by 2030, against 49 million tonnes in the previous year’s review. The pipeline shrank in a single year, and the review says so plainly.
The same review prints the other half of the picture. Production from projects that are operating or have reached a final investment decision is set to reach 4.2 million tonnes per year by 2030, a fivefold increase on 2024 production. Both statements are true at once. The announced pipeline is contracting while the committed base expands, and a reader who hears only one of them has been told half of the story.
The European Court of Auditors, looking at the European Union’s own hydrogen targets, reached a blunter conclusion: based on the information available from member states and industry, the targets turned out to be overly ambitious and are unlikely to be met by 2030. That is an audit judgement on targets. It measures no plant and says nothing about what is being built this year.
France offers a narrower and more concrete count. The French hydrogen industry body’s barometer, published in January, reports 50 MW of electrolysis capacity installed in France and 250 MW that has reached a final investment decision or is under construction. This is the body’s own count, not an official statistic, and it is printed here as exactly that.
Hydrogen: the ambition shrank, the committed base still grows
| Scope | Measure | What the document prints |
|---|---|---|
| World, announced for 2030 | Low-emissions output | 49 Mtpa a year earlier, 37 Mtpa now |
| World, operating or FID | Output by 2030 | 4.2 Mtpa, fivefold 2024 production |
| France, end of 2025 | Electrolysis capacity | 50 MW installed; 250 MW at FID or under construction |
| EU targets | Audit judgement | Unlikely to be met by 2030 |
Sources 3 International Energy Agency · 7 France Hydrogène · 4 European Court of Auditors
Read together, the four rows describe an industry leaving the announcement stage in smaller numbers than it entered it. The projects that remain are the ones with money committed. For the people who build and run them, that is a more stable place to work than a pipeline of intentions, even if it is a smaller one.
Battery recycling: waiting for batteries to wear out
Battery recycling is often described as a plant problem: build enough recycling capacity and the materials will follow. The documents describe a supply problem instead. The European Court of Auditors found that recycling covers on average just 10% of the demand for the materials concerned, an average for 2023 and not a current figure.
The International Energy Agency explains why that share cannot rise quickly. End-of-life batteries from electric vehicles and storage are expected to make up only one-third of recyclable material by 2030, with manufacturing scrap making up the rest, and it will take about a decade for end-of-life batteries to become the main source. The batteries that will feed recycling plants at scale are, for the most part, still in cars.
The French public debate commission provides the only plant-level document in this family. In a decision dated 5 February 2025, it took note of the suspension of ReLieVe, Eramet’s planned recycling plant for electric vehicle batteries in Dunkirk, which the company had notified in October 2024. The decision gives no reason and no capacity. It proves that one announced French plant stopped before construction. It does not prove that any plant is idle, and this piece does not say so.
Battery recycling waits for batteries to reach end of life
| Document | What it prints | Scope |
|---|---|---|
| European Court of Auditors | Recycling covers on average just 10% of material demand | EU, 2023 average |
| IEA | End-of-life batteries only one-third of feedstock by 2030; about a decade to become the main source | World, projection |
| CNDP | Takes note of the suspension of Eramet’s ReLieVe plant in Dunkirk | One French plant, 5 February 2025 |
Sources 5 European Court of Auditors · 6 International Energy Agency · 8 Commission Nationale du Débat Public
The honest reading is therefore about timing. The material that recycling needs arrives slowly, and a plant that is announced before its feedstock exists can be paused. For the process engineers who will run these plants, the question to ask of an employer is not how large the plant will be but where its material will come from, and when.
The counter-case
The strongest objection to this reading is that it is built on dates that have already passed. A dashboard closed at the end of one year, a release from one autumn and a barometer from one winter can all be overtaken, and a project that is only awarded in one document may be under construction in the next. That objection is correct, and it is the reason every row in this piece carries its date. The state of a project is a fact about a moment, not a permanent label.
A second objection concerns hydrogen. It is possible to read the shrinking pipeline as a collapse, and some coverage has. The review itself does not support that. The committed base is growing, and the auditors’ judgement is about targets, not about the plants that have reached a decision. This piece prints both halves because the documents do.
A third objection concerns recycling. One suspended plant is one plant. The decision cannot stand for the industry, and we do not ask it to; it stands only for the fact that an announced plant can stop, which is the point the feedstock projection makes in general terms.
What would change the reading
The reading would change if the statistics service reported offshore parks moving from permitting into operation faster than new projects enter review, if the next hydrogen review raised the announced pipeline again, or if a public record showed a French recycling plant running at scale on end-of-life batteries rather than scrap. Each of those would be a dated document, and each would move a project up a state in the tables above. We will update the piece when one appears, and record the change.
What this means for your career
The three families call for three kinds of work, and each simulation linked below practises one of them.
Offshore wind, in the state the documents describe, needs people who turn an award into a project: coordinating surveys, suppliers, the grid connection and the permits that are still under review. That is the work of an offshore wind farm project engineer.
Hydrogen, where money is committed, needs people who make electrolysers work in a real plant: sizing, integration, safety and the operating limits of the stack. That is the work of a green hydrogen engineer on PEM electrolysers.
Battery recycling, where feedstock is the constraint, needs process engineers who can run a line on the material that actually arrives, mostly manufacturing scrap for now, and plan for the material that will arrive later.
If you want to learn how an awarded park becomes a building site, start with the offshore wind project engineer simulation. If you want to work where committed money is turning into plants, start with the green hydrogen electrolyser engineer simulation. If you want to understand why a recycling plant depends on what comes through its gate, start with the lithium-ion battery recycling process engineer simulation. None of these documents tells you how many jobs there are, and we do not pretend otherwise. Each simulation lets you do the work before you choose it.