The Hidden Skill Gap in US Transit Engineering — and What to Do About It

The US transit industry has had a quiet workforce problem for as long as I have worked in it. The agencies and the vendors do not call it a skill gap publicly, because the language of skill gaps invites the wrong policy response, and because the agencies that admit it most loudly are the ones that have the hardest time recruiting against it. But every transit chief signal engineer, every vendor regional manager, and every consulting principal who staffs CBTC procurements in 2026 is operating inside the same unspoken constraint: there are not enough US engineers with the right combination of rail-domain knowledge, embedded-and-real-time systems experience, and safety-critical software literacy to staff the projects the agencies want to run. The post below is a practitioner’s account of where the gap actually lives, why it has persisted across two decades of growing demand, and the practical steps an agency, a vendor, or an individual engineer can take to address it.

The shape of the gap

The US transit engineering skill gap is not a single monolithic shortage. It is a layered shortage that hits different roles differently.

The shallowest layer is the entry-level shortage of new graduates with rail-relevant coursework. The five US universities with named rail engineering programs (UIUC, Michigan Tech, Penn State, Virginia Tech, San José State Mineta) graduate together a few hundred rail-relevant master’s and doctoral students per year. The total US transit signal engineering hiring demand — agencies and vendors combined — runs higher. The agencies and the vendors compete for the same pool, and most of them lose half their hires to mainline freight, aerospace, or general infrastructure. (For the underlying education map, see Top US Universities for Rail Engineering Education.)

The middle layer is the scarcity of mid-career engineers with one or more deployment cycles completed. Five years of relevant CBTC work — a deployment where the engineer ran factory acceptance, did a revenue cutover, investigated incidents, and held design responsibility for a subsystem — is the threshold at which an engineer is independently useful on a complex US procurement. The market for engineers who clear that threshold is tight. The supply is limited by the deployment cadence: the US has only a handful of CBTC retrofits in revenue service simultaneously, and the engineers who get the experience can be counted in low thousands. The agencies and the vendors compete for them aggressively, and the consulting firms that staff procurements compete for the same group.

The deepest layer is the senior-engineer scarcity. The US transit signal engineering community of senior engineers — twenty-plus years of relevant experience, ability to write a safety case, ability to lead an integration team across vendor and agency interfaces — is small. Many of the strongest senior engineers in this country are within ten years of retirement. The succession pipeline behind them is thinner than it should be, because the mid-career layer is itself thin, and the senior layer cannot grow faster than the mid-career layer feeds it.

Each layer has different drivers and different remedies. Treating the entire shortage as one problem produces one-size-fits-all programs that miss most of the actual gap.

Why the gap has persisted

Three structural reasons keep the gap open.

The first is the labor-market mismatch with software. Rail signaling work, particularly on the VOBC and Zone Controller side, is embedded-and-real-time software development for safety-critical infrastructure. The labor market that values that skill set most highly is software, where compensation is consistently 30 to 60 percent higher for equivalent experience. (For the salary picture, see CBTC Engineering Salary in the US: 2026 Benchmarks.) The transit industry has not been willing or able to close that gap, and many of the engineers most qualified to do safety-critical real-time software work choose general software employers instead.

The second is the project-cadence mismatch. CBTC deployments are decade-scale projects. An engineer who joins a vendor at the start of a US retrofit and stays for the full deployment will get one or two end-to-end project cycles in a fifteen-year career. The engineer who wants more variety leaves for adjacent industries — automotive, medical devices, aerospace — where the cycles are shorter, the compensation is higher, and the technical challenges overlap substantially. The transit industry’s deployment cadence does not lend itself to the rapid skill-building that early-career engineers often want.

The third is the institutional-memory bottleneck. Much of US transit engineering knowledge lives in the heads of individual senior engineers and in the procedures of individual agencies. The discipline does not have the same level of canonical textbooks, structured certification programs, or open-architecture documentation that European and Asian rail engineering communities have built. An engineer entering US transit signal engineering in 2026 has to reconstruct from scratch knowledge that should be sitting in published references. (The reading list in Top 10 Books and Courses Every Rail Signaling Engineer Should Read is the start of an answer to this.) The result is that the knowledge transfer from senior to junior is slower than it should be, and the agencies and vendors that lose senior engineers to retirement lose more than they realize.

The US transit engineering skill gap is layered. Each layer has different drivers and different remedies.

What the regulators and the safety community have noted

The skill gap has not been entirely invisible to the federal safety community. The National Transportation Safety Board, in its 2018 investigation report on the 2017 Washington Metropolitan Area Transit Authority derailment in the Red Line tunnel near Farragut North, identified inadequacies in technical staffing and signal-system maintenance practices as contributing factors. The NTSB has issued similar staffing-related findings in other transit accident investigations across the past two decades. The Federal Transit Administration’s Transit Worker Safety summit reports and the National Transit Database staffing surveys have flagged signaling and systems-engineering staff shortages at multiple agencies. The Transit Cooperative Research Program has published several reports on transit workforce development, including TCRP Reports on workforce planning and on succession-planning practices, that document the same picture from the agency-management perspective.

The findings are not headline material in the way that an accident itself is, and the regulatory response has been measured. The reality on the ground at most US transit agencies is that the staffing constraint is real, persistent, and worse than the public reports describe. The agencies that admit it most candidly are the ones with the most aggressive recruitment programs.

What agencies can do

Several practical steps are available to a US transit agency that wants to address the skill gap on its property.

Build internal training pipelines that do not depend on the vendor. Most US transit agencies rely on vendor-led training to bring junior engineers up to speed on the specific CBTC system they have purchased. That training is necessary but it is not sufficient, because it teaches the vendor’s product rather than the discipline. An agency that invests in a multi-year internal training pipeline — sponsored continuing education at AREMA and APTA, structured shadowing with senior signal engineers, sponsored master’s-level coursework at UIUC, Penn State, or Virginia Tech, and a clear technical-track promotion path — will retain engineers longer than an agency that does not.

Create technical career tracks that are genuinely competitive with managerial promotion. The US transit industry has historically rewarded engineers who became managers and underrewarded engineers who stayed in deep technical work. The vendors do this better than the agencies do, but neither does it as well as the software industry does. An agency that creates principal-engineer and chief-engineer roles with compensation that competes with management roles will keep more of its senior engineers from leaving for vendor or consulting employers.

Engage early with master’s and doctoral students. The named rail programs at UIUC, Michigan Tech, Penn State, and Virginia Tech are smaller than the agencies’ total hiring need, but the early-engagement pipeline that funds graduate research, hosts internships, and sponsors thesis projects is the most direct way to recruit graduates. The agencies that show up at TRB, that fund TCRP-aligned research, and that recruit at the named programs build a stronger pipeline than the agencies that do not.

Reduce the friction on lateral transfers from adjacent industries. Engineers in automotive embedded systems, in aerospace flight control, in medical device firmware, and in telecommunications infrastructure have most of the relevant skills for CBTC work. The friction is on the rail-domain side: the discipline-specific vocabulary, the standards literacy, the operational context. An agency that builds a structured lateral-transfer program — six months of paid training, a defined ramp-up project, an explicit mentor from the senior signal engineering staff — will recruit successfully from adjacent industries that the broader rail discipline has historically ignored.

What vendors can do

The vendor side of the skill gap is somewhat different. The vendors generally have more competitive compensation than the agencies and a deeper internal training program, but they face the same upstream supply constraint and they lose engineers to the same software-industry pull.

The vendors can invest in open documentation and open architecture more aggressively than they currently do. The current US CBTC vendor practice — proprietary protocols, closed source, opaque interface control documents — is consistent with vendor commercial interest but is also a major contributor to the discipline’s knowledge-transfer problem. A vendor that commits to fuller documentation, to publishable interface specifications, and to a more open posture on the non-safety-critical APIs makes the discipline easier for new engineers to enter and reduces its own training burden. The European EULYNX and OCORA initiatives provide a model that US vendors have so far been slow to follow.

Vendors can also fund university programs more directly than they currently do. The vendor relationships with US rail engineering programs are limited compared with the parallel relationships in Europe and Asia. A vendor that funds a named professorship, a graduate fellowship, or a sustained research program at one of the named US rail engineering universities builds a recruitment pipeline that pays back across a decade.

What an individual engineer can do

For an engineer trying to navigate the gap from inside it, the practical advice is simpler.

Read the standards. Most US engineers in the discipline have not read IEEE 1474 end to end, have not read IEC 62290 end to end, and have not read EN 50126/8/9 in detail. An engineer who has read them and can argue about specific clauses operates at a different level than one who has not.

Build cross-domain depth. The engineers most in demand at the senior level are the ones who can bridge two or more of the discipline’s silos: onboard and wayside, RF and embedded software, safety case and operational performance, vendor product and agency operations. Cross-domain depth is hard to build inside a single role; the engineer who deliberately rotates across project assignments, agencies, or vendor product lines builds it faster than the one who does not.

Show up where the discipline talks. The APTA Rail Conference, the IEEE/ASME Joint Rail Conference, the TRB Annual Meeting, and the AREMA annual meeting are the places where US transit signal engineering knowledge is shared in public. An engineer who attends, presents, and reads the proceedings is on a faster knowledge-acquisition trajectory than one who does not.

Consider the lateral move into rail. For an engineer in automotive, aerospace, medical, or telecom who is reading this post and wondering whether the move is worth making, the honest answer is yes, conditionally. The compensation may step down on the move; the work cycles are longer; the regulatory environment is more conservative. The compensating factor is a discipline that handles consequential public infrastructure, that has a long career runway, and that is genuinely short of capable engineers. (For the cross-discipline mapping, see How to Become a CBTC Engineer: A 2026 Career Path.)

What this means in practice

• The US transit engineering skill gap is layered: an entry-level shortage of rail-relevant graduates, a mid-career scarcity of engineers with one or more deployment cycles, and a senior-engineer succession risk as a generation approaches retirement. Each layer has different drivers and different remedies.
• The gap has persisted because of compensation mismatch with software, the long deployment cadence of US CBTC projects, and the institutional-memory bottleneck in a discipline that has under-invested in canonical references and structured knowledge transfer.
• Federal safety findings, FTA workforce reports, and TCRP studies have documented elements of the picture. The on-the-ground reality at most US transit agencies is more constrained than the public reports describe.
• Agencies can build internal training pipelines, create competitive technical tracks, engage with master’s and doctoral programs, and reduce the friction on lateral transfers. Vendors can invest in open documentation and in university programs.
• An individual engineer can read the standards, build cross-domain depth, attend the discipline’s conferences, and consider lateral moves. The discipline is genuinely short of capable engineers, and a deliberate plan compounds across a career.

Where to go next

This post is a 9-minute briefing. The full discussion of the workforce, lifecycle, and US deployment context lives in Communications-Based Train Control (Volume 2): Buy on Amazon.

Sources

• Wang, C. (2026). Communications-Based Train Control, Volume 2: Operations, Lifecycle & US Deployment. Independent. ISBN 979-8-258-54295-3.
• National Transportation Safety Board. Rail Accident Investigation Reports. ntsb.gov
• Federal Transit Administration. Transit Worker Safety Reports and National Transit Database. transit.dot.gov
• Transit Cooperative Research Program. TCRP Reports on Transit Workforce. Transportation Research Board. trb.org/TCRP
• American Public Transportation Association. Workforce Development Resources. apta.com
• US Bureau of Labor Statistics. Occupational Outlook Handbook — Electrical and Electronics Engineers. bls.gov/ooh
• American Railway Engineering and Maintenance-of-Way Association. Continuing Education and Conferences. arema.org
• IEEE Standards Association. IEEE Std 1474.1: Standard for Communications-Based Train Control (CBTC) Performance and Functional Requirements.