COOKED
The core of this job — turning engineer sketches and specs into dimensioned 3D models, 2D detail sheets, BOMs, and revision packages in SolidWorks/Inventor/Creo — is exactly the parametric, rule-bound screen work that CAD automation, model-based definition, and generative tooling are absorbing fastest. Drafters do not stamp drawings; a licensed PE or the responsible engineer signs, so there is no liability shield protecting the seat. What survives is thin and mostly tied to physical shops: reverse-engineering existing hardware with calipers and scanners, GD&T and tolerance-stack judgment that manufacturing actually trusts, and sitting between design intent and the machinist who has to cut it.
Part 2020 shock, part continued decline in the years since.
Median pay $57,060 → $71,550 +0.3% 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
-6.5% 42,900 → 40,100 on the projections basis
Exposed, and shrinking
Both signals point the same way: the tasks are largely automatable and the BLS projects -6.5% by 2034. This is the case where the score and the forecast agree, and it is the one worth taking seriously.
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.
~3,300 openings a year on average, including replacing people who leave.
DrafterDetailerDie DesignerBody DesignerGage DesignerMold DesignerTool DesignerDesign CheckerDesign DrafterDetail DrafterPatent DrafterPiping DrafterAutoCAD DrafterPattern DrafterPiping DesignerAutoCAD DesignerCastings DrafterProduct DesignerProject DesignerControls DesignerDesign TechnicianAutoCAD TechnicianFurniture DetailerMechanical Drafter
Holding it up: task resistance . Weakest point: liability shield .
Core tasks are already automatable Detailing a weldment from an engineer's markup — projecting views, applying dimension schemes, populating title blocks, exploding assemblies for BOMs, pushing an ECO through revision control — is deterministic work driven off an existing parametric model, and MBD plus automated drawing-creation tools already do the view generation and annotation extraction, which is why this sits at 6 rather than mid-band: the tasks aren't partially automatable, they're the demo case.
Fully desk- and screen-based The 4 covers the real but occasional trips out of the chair: taking calipers and a scanner to a legacy part with no drawing, walking a print out to the CNC operator to check a datum callout, red-lining on the shop floor during first-article — everything else happens on a dual-monitor workstation and can be done from anywhere with a VPN.
No licence, no signature requirement No state licenses mechanical drafters and no ASME standard requires a credentialed drafter's signature; the approval block is signed by the responsible engineer or PE, and ASME Y14.5 certification or an ADDA credential helps you get hired but transfers no legal exposure to you, so the 2 reflects credentials that are resume items, not statutory gatekeeping.
Executes defined procedures on defined inputs Choosing a tolerance stack, deciding which surface becomes the primary datum, and flagging that a called-out fillet won't clear the cutter are genuine calls that keep this off the floor, but they resolve against Y14.5, company drafting standards, and an engineer's review before release — you raise the issue, someone else owns the decision, which is a 6 rather than a 10.
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 (6/20 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 (2/20) is whether the law requires a licensed human to sign. Trust premium (3/20) is whether buyers specifically pay for a person. Judgment and accountability (6/20) 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 11 of this occupation's 21 points (52%).
Embodiment (4/20) 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.
Aerospace Engineers 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 36/100 — EXPOSED.
Task-mix shift toward GD&T authoring and tolerance-stack analysis: if generative CAD absorbs the modeling and sheet-generation tier, the remaining seat is ASME Y14.5 datum scheme definition, tolerance allocation against actual process capability, and MBD/PMI validation that shops will accept — work where an error scraps a production run and where checkers currently reject a large share of AI-generated PMI. Watch ASME Y14.5-certification (GDTP) becoming a de facto hiring line rather than a nice-to-have.
Consolidation of the role into shop-floor reverse engineering and first-article work: portable CMM/Faro-arm and structured-light scanning of legacy hardware with no drawings, scan-to-CAD cleanup, and walking the part with the machinist. Defense and MRO sustainment programs (legacy aircraft, Navy shipyard parts with lost tech data packages) fund exactly this. Raises only if the seat is redefined around the metrology hardware rather than the workstation.
Adoption of a formal checker-of-record signature block in the quality system — not a PE stamp, but AS9100/ISO 9001 and DO-254/NADCAP-style audit requirements that a named, qualified human verify and sign released drawings and MBD datasets. If FAA or DoD contracting language (e.g., MIL-STD-31000B technical data package deliverables) is amended to require a named human verifier for AI-generated model-based definition, the signature becomes contractual rather than merely procedural.
Ownership of the released-configuration decision: as generative tools produce many candidate models, the consequential call becomes which revision is released, what change-order impact it carries downstream to tooling and inspection fixtures, and whether a deviation is accepted. If CCB (configuration control board) membership and ECO authority formally sit with the drafting/checking seat rather than the design engineer, this rises.
The limit. Even with every lever, this occupation is structurally capped: no licensing board exists for drafters, no buyer pays for a human-drawn detail sheet, and the levers described mostly convert the job into a smaller number of checker/metrology/configuration seats rather than preserving headcount. Trust premium has no plausible route — the customer buys the part, not the drawing.
| Dallas-Fort Worth-Arlington, TX | 1,620 | $83,200 +16% |
| Los Angeles-Long Beach-Anaheim, CA | 1,480 | $80,780 +13% |
| Houston-Pasadena-The Woodlands, TX | 1,300 | $78,820 +10% |
| New York-Newark-Jersey City, NY-NJ | 1,180 | $84,070 +17% |
| Seattle-Tacoma-Bellevue, WA | 1,020 | $94,470 +32% |
| Detroit-Warren-Dearborn, MI | 970 | $75,050 +5% |
| Minneapolis-St. Paul-Bloomington, MN-WI | 830 | $75,770 +6% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 730 | $74,100 +4% |
| Boulder, CO | 60 | $102,710 +44% |
| San Francisco-Oakland-Fremont, CA | 390 | $98,310 +37% |
| San Jose-Sunnyvale-Santa Clara, CA | 240 | $97,450 +36% |
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 21. 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.