EXPOSED
Operating tampers, ballast regulators, spike drivers and tie-inserters on live right-of-way in weather, at night, around traffic and shifting geometry is about as far from text-and-screen work as jobs get — language AI touches almost none of it. The real exposure is mechanical, not cognitive: automated tamping/lining machines with onboard measurement systems and self-propelled gangs let fewer operators cover more track-miles, so the occupation thins rather than disappears. FRA roadway worker protection rules, track inspection sign-offs and BMWED/SMART agreements keep a certified human on the ground making foul-time and slow-order calls.
Headcount grew steadily across the period.
Median pay $56,100 → $70,070 -0.1% 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
+1.6%
Percentage only. The projection counts a different population from the 19,580 above — it includes self-employed workers, which for this occupation is most of them, so the two headcounts are not comparable.
Growing, and only partly exposed
The BLS expects +1.6% more of these jobs by 2034, and at 63/100 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.
~1,100 openings a year on average, including replacing people who leave.
TrackmanTrack ManTrack LayerTrackwalkerSection HandTrack WalkerTrack WelderTrack WorkerTrack DresserTrack LaborerTrack MechanicTrack RepairerSection LaborerTrack InspectorMachine OperatorRail Track LayerTrack MaintainerMaintenance LaborerTrack Repair PersonTrack Repair WorkerTrack Service PersonTrack Service WorkerRail Track MaintainerOil Distributor Tender
Holding it up: embodiment . Weakest point: trust premium .
Tasks largely resist digitisation Setting spike, seating ties, running a tamper's workheads into a turnout frog where the machine's automatic mode gives up, and rerailing a derailed piece of equipment are physical judgment jobs no language model or fixed script performs — the 17 rather than 20 reflects that automatic tamping/lining cycles with laser and chord measurement already do the plain-track surfacing that used to take a full gang.
Hands-on in uncontrolled environments You are on ballast in the rain, at 0200 under work lights, with hi-rail traffic behind you and 39-foot rail being handled by a Kershaw boom — a 19 covers uncontrolled outdoor grade, live right-of-way, and gauntlet/bridge clearances that leave no room to step back, short of the underground or over-water extremes that would earn 20.
Certification preferred, not legally required Roadway worker protection and on-track safety qualification under 49 CFR 214, plus carrier-specific machine operator certifications and annual recerts, are real gates on who touches the controls — but the licence sits with the railroad's program and the track inspector signs the geometry, so an operator carries discipline and decertification risk rather than the personal professional liability a licensed engineer holds, which puts it at 8 rather than 14.
Meaningful discretion Calling foul time, deciding a joint is bad enough to slow-order, judging when ballast is short before you release track to traffic, and reading whether a lift will hold in soft subgrade are consequential unscripted calls made in minutes — a 13 rather than 16 because dispatcher authority, the FRA class limits and the foreman's final say bound almost every one of them.
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 (17/20 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 (8/20) is whether the law requires a licensed human to sign. Trust premium (6/20) is whether buyers specifically pay for a person. Judgment and accountability (13/20) 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 27 of this occupation's 63 points (43%).
Embodiment (19/20) 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.
Ship Engineers SAFE
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 77/100 — SAFE.
Task-mix shift: as onboard measurement and self-propelled production gangs absorb the repetitive surfacing/lining passes, the remaining role concentrates on foul-time negotiation with the dispatcher, thermal-stress and rail-neutral-temperature decisions, unstable subgrade calls, and derailment-risk judgment under time pressure with traffic waiting. Watch for BMWED/SMART-MWED contract language creating a distinct higher-rated RWIC or machine-operator-in-charge classification separate from the machine operator seat.
FRA rulemaking under 49 CFR Part 213/214 requiring a named, certified roadway worker in charge (RWIC) to personally sign off on post-tamping track geometry restoration and slow-order removal before traffic resumes — i.e. extending the existing inspector certification regime to machine-produced surfacing work rather than accepting onboard measurement records as the record of compliance. The FRA's automated track inspection waiver fights (BMWED petitions against BNSF/NS ATIP waivers, 2022-2024) are the live venue: if the agency rules that machine data cannot substitute for a certified human sign-off, this rises.
State-level two-person crew and roadway-worker-protection statutes (following the pattern of the freight two-person crew laws passed in Nevada, Arizona, Ohio, Kansas, Washington and the FRA's 2024 crew size final rule) extended to maintenance-of-way gangs, mandating a minimum certified lookout/RWIC per machine consist regardless of onboard automation.
Post-derailment regulatory tightening (East Palestine follow-on: NTSB recommendations and the Railway Safety Act reintroductions) mandating physical hands-on verification of joint bars, welds, and fastener condition at intervals machine-based geometry cars cannot satisfy — pushing work back toward on-ground human tasks that automated gangs skip.
The limit. Trust premium has no realistic route: the buyer is a handful of Class I railroads and transit agencies buying track-miles per shift, and none of them will pay more for human-operated tamping. The dominant force here is capital intensity, not capability — a lever that raises liability_shield keeps a certified human present per machine but does not stop headcount thinning as one consist replaces three gangs. task_resistance and embodiment are already near ceiling and cannot move much.
| New York-Newark-Jersey City, NY-NJ | 2,710 | $87,150 +24% |
| Boston-Cambridge-Newton, MA-NH | 280 | $85,240 +22% |
| Chicago-Naperville-Elgin, IL-IN | 220 | $60,120 -14% |
| Houston-Pasadena-The Woodlands, TX | 180 | $52,900 -25% |
| Pittsburgh, PA | 160 | $59,390 -15% |
| Riverside-San Bernardino-Ontario, CA | 130 | $65,060 -7% |
| Dallas-Fort Worth-Arlington, TX | 100 | $56,470 -19% |
| Longview, TX | 90 | $50,140 -28% |
| Baltimore-Columbia-Towson, MD | 50 | $87,410 +25% |
| New York-Newark-Jersey City, NY-NJ | 2,710 | $87,150 +24% |
| Boston-Cambridge-Newton, MA-NH | 280 | $85,240 +22% |
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 63. 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.