EXPOSED
Much of the day is CAD modeling, hydrostatic and stability calculations, weight estimates, resistance/powerplant sizing, specification writing and class-rule compliance checks — all of which are increasingly generated or pre-checked by parametric tools plus AI, and naval architecture already runs on heavy computational automation. What resists is the physical side: shipyard construction oversight, sea trials, damage surveys, machinery troubleshooting aboard vessels, and the signed engineering judgment that a hull or propulsion arrangement is safe and class-approvable. Modal worker here is a design-office engineer at a shipyard, design firm, or Navy/NAVSEA organization, where PE stamps, ABS/USCG approval submittals, and personal accountability for safety cases anchor the role.
Nearly all of this fall was the 2020 shock. It has been climbing back since.
Median pay $92,400 → $112,230 -2.8% in real terms
This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.
So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.
BLS projection, 2024–2034
+5.8% 8,500 → 9,000 on the projections basis
Growing, and only partly exposed
The BLS expects +5.8% more of these jobs by 2034, and at 54/100 the work is only partly exposed — some tasks are automatable, the core of the job is not. Nothing here is in tension.
Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.
~600 openings a year on average, including replacing people who leave.
EngineerBoat DesignerShip SurveyorNaval EngineerMarine EngineerMarine SurveyorNaval ArchitectPiping DesignerMarine ArchitectProject EngineerMarine ConsultantAutomation EngineerNavigation EngineerStructural DesignerStructural EngineerArchitect SpecialistMarine Civil EngineerNaval Marine EngineerEngineering SpecialistMarine Design EngineerMarine Service EngineerShip Equipment EngineerMarine Equipment EngineerConsulting Marine Engineer
Holding it up: judgment & accountability . Weakest point: trust premium .
Mixed — a routine tier and a judgment tier Hull form optimization, hydrostatic and intact/damage stability runs, resistance-powering estimates and weight books are already parametric and largely tool-generated, but shipyard build oversight, sea trial instrumentation and troubleshooting, docking surveys and reverse-engineering legacy machinery on ships with no as-built drawings still require a person on the plating — a roughly even split, which is why it sits at 11 and not in the resistant band with field-only engineering roles.
Some physical or field component Most weeks are at a workstation in a design office or NAVSEA code, but the job periodically puts you in drydock, in tanks and voids during survey, on inclining experiments and on builder's/acceptance trials at sea — real confined-space and shipboard exposure, though not the daily uncontrolled-environment work of a marine surveyor or engineering officer, hence 10 rather than 15.
Licensed human required and personally liable Approval submittals to ABS/DNV and USCG, and structural or stability calculations released under a PE stamp, place named personal accountability on an individual engineer, but a large share of the workforce — federal civilian naval architects and shipyard design staff working under an approved quality system and industrial exemptions — practices without a licence, which pulls this to 11 instead of the 16+ of professions where no licence means no work.
Exists to be accountable for ambiguous calls You decide whether a damaged-stability case is acceptable, whether a weld or scantling deficiency found during construction is a repair or a redesign, what margins to carry on a first-in-class hull, and whether a vessel is fit for trials — calls made with incomplete data where being wrong sinks ships, which is why this is 14 even though the rulebooks (SOLAS, class rules, Subchapter requirements) constrain the envelope.
The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.
Your task mix speaks to task resistance (11/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (11/20) is whether the law requires a licensed human to sign. Trust premium (8/20) is whether buyers specifically pay for a person. Judgment and accountability (14/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 33 of this occupation's 54 points (61%).
Embodiment (10/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
Construction Managers EXPOSED
The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 69/100 — SAFE.
Task-mix shift: this occupation genuinely has two tiers. If parametric/AI tooling absorbs the CAD production, weight books, resistance estimates and rule-check passes, the residual role concentrates on novel-arrangement safety cases, alternative-design-and-arrangement submittals under SOLAS II-2/17, ammonia and hydrogen fuel risk assessments under IGF Code alternative design approval, and reconciling as-built deviation with approved drawings — work with no verified precedent corpus.
Naval and commercial adoption of ammonia, methanol, hydrogen and battery-hybrid propulsion places consequential calls (bunkering arrangement, toxic-release siting, thermal-runaway containment) under genuine ambiguity with thin rule coverage, so the engineer owns the risk-acceptance decision rather than applying a rule.
USCG/ABS explicitly requiring a named licensed PE (or ABS-recognized responsible engineer) signature and traceable human review on stability letters, damage-stability and structural calculations where any part was machine-generated — the Subchapter M towing-vessel and Subchapter L small-passenger-vessel plan-review regimes already route approvals through named submitters, and the post-Conception (SUV) NPRMs tightened plan-approval accountability. A parallel exists in NAVSEA's Technical Warrant Holder system, where a single named warrant holder personally owns a technical authority decision; extending warrant-holder countersignature to AI-derived design products would formalize it.
If shipyard schedule pressure and the shortage of surveyors keep pushing design engineers into on-site duty — construction oversight, weld and alignment disputes, inclining experiments, sea trials, damage surveys and grounding/allision investigations — the share of the day that is unpredictable-environment physical work rises. Class-society remote-survey expansion pushes the other way, so this only holds if remote survey stays limited to routine periodic items.
Insurer/P&I club and class-society conditions on cover requiring documented licensed-engineer verification of software-generated hull girder and stability outputs after a casualty traced to an unverified automated calculation — IACS UR E26/E27 cyber rules show class already willing to write software-provenance requirements into class notation.
The limit. No plausible route to a higher trust_premium: buyers are shipyards, owners and government programs procuring on price, schedule and class approval — none of them pay a premium for human authorship as such, only for a signature that satisfies a regulator or underwriter. That value belongs in liability_shield, not trust. Also note liability_shield and task_resistance here partly trade off: the same class-society digitalization that could mandate human countersignature is also building remote survey and automated rule-check pipelines.
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 710 | $159,700 +42% |
| New York-Newark-Jersey City, NY-NJ | 250 | $112,230 +0% |
| Portland-South Portland, ME | 230 | $104,850 -7% |
| Bremerton-Silverdale-Port Orchard, WA | 180 | $106,250 -5% |
| Seattle-Tacoma-Bellevue, WA | 150 | $128,610 +15% |
| Barnstable Town, MA | 120 | — |
| Urban Honolulu, HI | 120 | $108,200 -4% |
| San Diego-Chula Vista-Carlsbad, CA | 90 | $118,100 +5% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 710 | $159,700 +42% |
| San Francisco-Oakland-Fremont, CA | 50 | $140,520 +25% |
| Seattle-Tacoma-Bellevue, WA | 150 | $128,610 +15% |
We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.
That is worth saying out loud next to a score of 54. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.