← Risk register SOC 17-2072 · reviewed 2026-08-11

Electronics Engineers, Except Computer

96,900 US workers · median $130,220/yr · Engineering

EXPOSED

The desk half of this job — datasheet trawling, block diagrams, component selection, SPICE setup, register maps, test plans, compliance paperwork, and glue firmware — is exactly what current models do at usable quality, and EDA vendors are shipping AI layout and verification assistants into the tools. The half that holds is bench reality: an analog front end that oscillates, EMI that only shows up in the chamber, a thermal problem in the enclosure, a board that passes sim and fails at -40C. Licensure is a weak moat here because the industrial exemption means most electronics engineers never stamp anything, so protection comes from debug skill and owning safety-critical design calls rather than from a seal.

10-year outlook: Headcount per product drops as AI absorbs schematic, firmware, and documentation work, while demand shifts toward engineers who can debug real hardware and own safety and manufacturability sign-off.

US employment, 2019–2025-24.8%
128,80096,900 workers

Part 2020 shock, part continued decline in the years since.

Median pay $105,570 → $130,220 -1.3% in real terms (nominal +23.3%, less ~25% US inflation over the period)

The job count is not the verdict

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

+6.2% 95,900 → 101,800 on the projections basis

Growing, and only partly exposed

The BLS expects +6.2% more of these jobs by 2034, and at 46/100 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.

~5,700 openings a year on average, including replacing people who leave.

One email if this score changes. Watch as many occupations as you like from the same address — no account, and nothing is sent on a schedule, only when a verdict actually moves.

Also known as — 24 job titles this covers

Titles reported by people doing this work, from the US Department of Labor's O*NET survey. If your job title is here, this page is about your work even though the name doesn't match.

EngineerTest EngineerField EngineerRadar EngineerDesign EngineerSignal EngineerSystem EngineerAntenna EngineerCircuit DesignerProduct EngineerWireless EngineerEquipment EngineerMicrowave EngineerAutomation EngineerCommercial EngineerDeployment EngineerElectrical EngineerEvaluation EngineerAudio Visual ManagerDevice Test EngineerTransmitter EngineerElectrophonic EngineerField Service EngineerOutside Plant Engineer

Score — 46/100 resistance

Holding it up: judgment & accountability (13/20). Weakest point: liability shield (6/20).

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 11 + 9 + 6 + 7 + 13 = 46. · Scored 2026-08-11, and re-examined when evidence accumulates rather than on a schedule.

Task resistance 11/20

Mixed — a routine tier and a judgment tier Schematic capture from a known topology, BOM selection against datasheet parameters, SPICE sweeps, register-map documentation, and DO-160/61000 test-plan drafting are already being generated at usable quality, but bringing up a switching supply that ripples at 300kHz, chasing a ground-loop that only appears with the motor running, or correlating a failed radiated-emissions scan to a specific clock harmonic still takes someone at the bench with a scope probe — that split is why this sits at 11 rather than in the resistant band.

Embodiment 9/20

Some physical or field component You are not in a trench, but you are also not screen-only: probing a live board, reworking 0402s and BGAs with hot air, sitting in the anechoic chamber with a spectrum analyzer, running thermal chambers to -40C and 85C, and crawling into an enclosure to find where the harness couples noise are recurring days, and that intermittent-but-real hands-on load is a 9 rather than a 15 because the environment is a lab you control, not a rooftop or a refinery.

Liability shield 6/20

Certification preferred, not legally required The industrial exemption in nearly every state PE statute means the vast majority of electronics engineers design, release, and ship products without ever stamping a drawing, so the enforceable barrier is your employer's design-review process and standards like IEC 60601 or ISO 26262 — the 6 credits the fact that safety-critical and utility-adjacent work does occasionally demand a licensed signatory, not that most of you carry one.

Trust premium 7/20

Some relationship component Your output is a board, a spec, and a test report that has to work regardless of who authored it, and the customer usually never meets you; the 7 reflects the real but bounded relationship capital with the fab, the contract manufacturer's process engineer, and the certification lab that gets your unit re-tested quickly rather than queued for six weeks.

Judgment & accountability 13/20

Meaningful discretion Deciding a derating margin, accepting a part with a 20-year EOL risk, signing off that an isolation barrier meets creepage under pollution degree 2, or calling a field-failure root cause fixed are decisions with recall-scale consequences and no procedure that settles them, which puts this at the top of the discretion band — short of 14+ only because those calls typically pass through a design review board rather than resting on you alone.

Confidence: medium · reviewed 2026-08-11 · how scoring works

What this job involves — and which parts are yours

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.

AI already does these at usable quality

These still need a person

Active moats on the surviving side: judgment, embodiment

How to future-proof this job

Training paths for your skill gaps: Khan Academy — physics, chemistry and biology from the ground up free · Khan Academy — mathematics, arithmetic through calculus free · MIT OpenCourseWare — operations management free · MIT OpenCourseWare — full course materials across every department, free free · Coursera — work planning and personal productivity free to audit · MIT OpenCourseWare — finance and accounting free · Coursera — project coordination and cross-team delivery free to audit

All 35 skills ranked by how many jobs they open →

Where this experience transfers — occupations you could move toward

Computed from U.S. Dept. of Labor O*NET skill and knowledge profiles: high overlap with what you already do, a materially higher resistance score, no large jump in required training, and no licence you would have to start a new pipeline to get. Targets that pay meaningfully less, that are themselves COOKED, or whose own headcount is falling are excluded — a move into a shrinking trade is not an escape.

Nuclear Engineers EXPOSED · 57/100 · you already have ~68% of the skill profile

Skills to close: Science, Mathematics, Operations Analysis, Active Learning

Architectural and Engineering Managers EXPOSED · 56/100 · you already have ~62% of the skill profile

Skills to close: Time Management, Management of Financial Resources, Active Learning, Coordination

What would move this back up — beyond any one person

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 66/100, still EXPOSED.

6 specific changes that would raise this score
  • already happening judgment accountability +3

    EU Cyber Resilience Act and RED Article 3(3) cybersecurity obligations force a named responsible person behind the Declaration of Conformity for connected hardware; if firms designate the lead electronics engineer as the accountable signer for security-relevant hardware design decisions (secure boot, key storage, debug-port lockdown), the role formally owns consequential calls it currently makes informally.

  • already happening task resistance +3

    Task-mix shift: this occupation genuinely has two tiers. If EDA AI assistants absorb schematic capture, component selection, register maps, SPICE setup and testbench generation, the residual job concentrates in the tier models do worst — root-causing an oscillating analog loop, chasing EMI that only appears in the anechoic chamber, thermal derating in a real enclosure, corner-case failures at -40C, and second-source respins under allocation. Resistance rises only if firms keep those tasks with the same job title rather than splitting off a separate 'hardware validation' occupation.

  • plausible liability shield +5

    Functional-safety and airworthiness regimes attaching personal, named sign-off to electronics design: DO-254 avionics hardware approvals via FAA Designated Engineering Representatives, ISO 26262 / IEC 61508 'competent person' safety-case signature, and FDA design-history-file review signatures on Class II/III device electronics. If auditors and notified bodies begin requiring that AI-generated schematics, FPGA/ASIC logic, and verification evidence be attested by a named engineer who accepts personal responsibility for the tool qualification (DO-330-style), the shield rises materially for the safety-critical slice of the occupation.

  • plausible embodiment +3

    Role absorption of lab and chamber work: if headcount cuts on the desk tier leave the remaining engineers personally running EMC pre-compliance sweeps, thermal chambers, HALT/HASS, oscilloscope/VNA probing on live boards, and field failure teardowns, the physical share of the day rises. This is a redefinition of the job's day, not a new capability barrier, so the gain is modest.

  • plausible trust premium +2

    Provenance requirements in defense and space: ITAR/DFARS and NASA/DoD contract clauses requiring documented human authorship and traceability of design artifacts, plus prohibitions on submitting technical data to non-accredited AI tools, would make a human-authored design a purchased attribute rather than a preference. Narrow — applies to cleared programs, not commercial consumer electronics.

  • unlikely liability shield +4

    Narrowing of the state industrial exemption. NCEES and several state boards have repeatedly floated limiting the manufacturing exemption; a state adopting PE-stamp requirements for electronics in life-safety systems (fire alarm, medical, utility-connected power electronics, EV charging infrastructure) would create an actual seal where today there is none. Watch state licensure bills and NFPA 72 fire-alarm design-document rules, which already push toward stamped electrical design.

The limit. The industrial exemption is the binding constraint: absent it, no seal exists for most of these 96,900 workers, and the liability levers above only reach the safety-critical minority (avionics, medical, automotive, utility). Commercial and consumer electronics design has no realistic route to a trust premium — buyers pay for a working, cheap board, not a human-drawn one. Realistic ceiling for the occupation as a whole is the high 50s; individual safety-critical niches can go considerably higher while the median consumer-hardware role does not move.

These are conditions, not forecasts — what would have to happen, not what will. Specific rules, cases and bills are named so you can go and check whether they exist and where they stand; verify before relying on any of them. Nothing here is legal or financial advice.

Where this work is, and what it pays there

BLS metro figures for 164 areas. The verdict above does not change by city — the rubric judges what the work involves, not where it happens — but pay and headcount do, and the national median hides a very wide range.

Most of these jobs

San Jose-Sunnyvale-Santa Clara, CA 5,970 $194,370 +49%
Boston-Cambridge-Newton, MA-NH 5,460 $142,920 +10%
Los Angeles-Long Beach-Anaheim, CA 4,880 $140,530 +8%
Dallas-Fort Worth-Arlington, TX 4,200 $135,150 +4%
Denver-Aurora-Centennial, CO 3,810 $127,260 -2%
San Diego-Chula Vista-Carlsbad, CA 3,450 $156,480 +20%
Washington-Arlington-Alexandria, DC-VA-MD-WV 2,690 $140,000 +8%
Detroit-Warren-Dearborn, MI 2,310 $128,650 -1%

Best paid

San Jose-Sunnyvale-Santa Clara, CA 5,970 $194,370 +49%
Santa Cruz-Watsonville, CA 30 $182,840 +40%
Santa Rosa-Petaluma, CA 290 $172,290 +32%

Percentages are against this occupation's national median of $130,220. Counts are jobs in that metro, not vacancies. Metros where the BLS suppressed the cell are absent rather than shown as zero.

Who is actually doing this — nobody, on the record

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 46. 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.

Read that as a gap in the reporting we can see, not proof of absence — the dispatch runs on English-language feeds and misses plenty. If you know of a case, tell us, or add a field report from inside the job.

Quick take — do you do this job?

Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

Field reports — what people say has changed

No field reports yet. A written account takes a paragraph rather than a tap, goes to an editor before it appears, and is the one thing on this page the rubric cannot produce on its own.

File a field report

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.

Watch this verdict
Kept current

Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.