Engine and Other Machine Assemblers
51-2031.00Median wage $53,710/yr34,000 employed (US)Rank #279 of 923 scored · top 30% by substitution
Construct, assemble, or rebuild machines, such as engines, turbines, and similar equipment used in such industries as construction, extraction, textiles, and paper manufacturing.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution34
Exposure30
Augmentation41
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
12 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
0%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%31
panel mean rating 2.2/5 → substitution pressure 31/100
Technical feasibility todayw 20%28
panel mean rating 2.1/5 → substitution pressure 28/100
Cost vs. human wagew 15%26
panel mean rating 2.0/5 → substitution pressure 26/100
Adoption barriersw 20%inverted — strong barriers lower the score55
panel mean rating 2.8/5 (barrier strength) → substitution pressure 55/100
Sector adoption velocityw 10%31
panel mean rating 2.2/5 → substitution pressure 31/100
Task breakdown (12 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Assemble systems of gears by aligning and meshing gears in gearboxes.
59CI 35–84 · exposure 58 · augmentation 25 · importance 4.0/5 · click for rater detail
Assemble systems of gears by aligning and meshing gears in gearboxes.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Automotive, industrial machinery, and precision equipment manufacturers have rapidly adopted robotic gear assembly over the past decade as technology matured. This reflects the digitized, capital-intensive manufacturing sectors where this task predominates. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing assembly automation adoption is steady but slow-moving compared to information-sector AI adoption, with robotics deployment concentrated in large-scale, high-volume production lines rather than broad diffusion. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Gear meshing and alignment is primarily a mechanical positioning task where AI/robotics offers little advantage to a human performing it; the value is in full automation rather than human-AI collaboration. Augmentation potential is limited because the task is not knowledge-work or judgment-heavy. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-driven vision systems and torque/alignment sensors can assist assemblers with quality checks and guidance, but this offers only incremental support rather than transformative productivity gains for the physical meshing task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Gear assembly is a highly structured, repetitive mechanical task with well-defined tolerances and clearance requirements. Computer vision and robotic arms with feedback control can reliably align, position, and mesh gears end-to-end, achieving significant time savings over manual assembly. |
| Task automatability | claude-sonnet-5 | 2/5 | Precise mechanical assembly involving tactile feedback, alignment, and meshing of physical gears requires robotic manipulation that is not yet reliably automatable end-to-end across varied gearbox designs with current general-purpose AI systems.rialize.co |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing requirement, liability asymmetry, or regulatory mandate necessitates human assembly of gears; the main barriers are capital investment and incumbent workforce transitions rather than legal/organizational ones. These are adoptable obstacles, not hard blockers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for this task, but quality/safety tolerances in gear meshing (e.g., aerospace, automotive) create liability and precision barriers that favor human dexterity or validated automated systems with extensive qualification. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated gear assembly equipment (capital + maintenance + energy) amortized over high-volume production is substantially cheaper per unit than skilled manual labor, though initial capital investment is significant. Operating costs per task strongly favor automation in any medium-to-high throughput scenario. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic assembly cells for gear systems require significant capital investment, custom tooling, and programming, making them cost-effective only at high volumes; for many contexts human assemblers remain cheaper or comparable. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Robotic assembly cells for gear systems are deployed in automotive and industrial manufacturing today, though integration complexity and part variability sometimes require human oversight. Production systems exist and perform this task at scale with high reliability, though occasional setup and tuning is needed. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Fixed automation and specialized robotic gearbox assembly lines exist in high-volume manufacturing (e.g., automotive), but flexible AI-driven assembly for varied or low-volume gear systems is largely research-stage or requires heavy custom engineering. |
Lay out and drill, ream, tap, or cut parts for assembly.
56CI 16–95 · exposure 50 · augmentation 25 · importance 4.1/5 · click for rater detail
Lay out and drill, ream, tap, or cut parts for assembly.
56| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Engine and machinery assembly has been heavily automated for decades; CNC drilling, reaming, and tapping are standard in major manufacturing sectors. Adoption is deep and mature in high-volume production environments. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/machining sectors adopt automation steadily but selectively, favoring dedicated CNC lines over general-purpose AI-driven layout work, so adoption for this specific flexible task remains slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Precision drilling and cutting tasks offer limited augmentation potential once automated; AI does not meaningfully assist a human performing these operations in concert. The task is suited for full replacement rather than human-in-the-loop assistance. |
| Augmentation potential | claude-sonnet-5 | 2/5 | CAD/CAM software and simulation tools can assist with layout planning and toolpath generation, offering some productivity gains, but the physical execution still relies heavily on human skill and machine operation. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Precision drilling, reaming, tapping, and cutting are operations well-suited to CNC machines and robotic arms controlled by CAM software. Current industrial automation systems can perform these tasks end-to-end with layout, tolerancing, and quality verification, delivering >50% time savings versus manual work at equal or superior quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical fabrication task requiring precise manual layout, drilling, reaming, tapping, and cutting of metal parts, which current AI systems cannot perform end-to-end without robotic hardware highly specialized for this exact task and part variability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Physical machines require capital investment and facility constraints, but there are no licensing requirements, regulatory prohibitions on automation, or legal mandates for human contact. Adoption is primarily economic and logistical rather than legal. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement exists, but quality/safety tolerances, liability for defective parts, and the need for skilled judgment in setup create meaningful organizational friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automation amortization per part is orders of magnitude lower than loaded human wages for the same precision drilling, reaming, and tapping work, especially in high-volume assembly. Even accounting for setup and programming, per-unit cost is negligible. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic/CNC automation can be cheaper per unit in high-volume dedicated lines, but for the flexible, low-volume, varied-part layout work typical of assemblers, capital and integration costs exceed human labor costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | CNC mills, drilling centers, and robotic work cells performing these exact operations are deployed in production at massive scale in automotive, aerospace, and machinery manufacturing. These are mature, proven technologies with documented reliability and precision far exceeding manual assembly. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | While CNC machining and some robotic drilling exist for high-volume standardized production, no generally deployed AI/robotic product performs flexible layout-and-machining of varied assembler parts reliably across diverse job-shop conditions. |
Read and interpret assembly blueprints or specifications manuals, and plan assembly or building operations.
39CI 30–48 · exposure 38 · augmentation 63 · importance 4.7/5 · click for rater detail
Read and interpret assembly blueprints or specifications manuals, and plan assembly or building operations.
39| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing remains relatively conservative in AI adoption; most assembly shops are small-to-medium firms with legacy systems and low digital infrastructure. While large OEMs experiment with AI-assisted planning, widespread production deployment is still limited and adoption remains slower than in information-intensive sectors. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and machine assembly are historically slower to adopt AI compared to information-sector work, with automation more focused on physical robotics than blueprint interpretation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist by rapidly generating and visualizing candidate assembly sequences, flagging potential conflicts, and cross-referencing specifications—helping planners work faster and catch errors. A human assembler or engineer remains in control and can override, but productivity gains are substantial. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can assist by summarizing specs, flagging inconsistencies, or answering technical queries, improving assembler efficiency without replacing the core planning judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Current AI can read and interpret technical drawings and generate assembly plans with reasonable accuracy, but real-world blueprints often contain ambiguities, legacy notation, or context-dependent decisions requiring domain expertise. AI could automate roughly half the task (document parsing, standard sequence generation) but human review for non-standard configurations remains necessary. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help interpret text-based specs, but reading complex mechanical blueprints and translating them into physical assembly plans requires spatial reasoning and integration with hands-on shop-floor context that current systems don't reliably handle end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Assembly planning usually requires human sign-off for safety and liability reasons, though not always by a licensed professional. Organizational processes often mandate engineer or lead-assembler review of plan feasibility, creating friction that slows full automation, though not a hard legal barrier. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement blocks AI use, but organizational reliance on experienced assemblers' tacit knowledge and liability for assembly errors create meaningful friction against full delegation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI inference for document analysis and plan generation is cheap, but integration with existing CAD systems and manufacturing planning tools, plus required human verification, brings total cost close to what a skilled assembler or planner would charge for the same work. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Specialized AI blueprint-reading tools require significant integration and human verification, so all-in costs remain comparable to or higher than skilled assembler labor for this planning task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Document AI and vision models can extract information from blueprints in controlled settings, and some CAD-linked planning tools exist, but production systems still have material error rates on complex or degraded prints. Deployed solutions exist but typically require significant human oversight and are not yet fully autonomous across the variety of real shop-floor blueprints. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some CAD-integrated and vision-language tools can parse technical drawings, but no deployed product reliably plans full assembly operations from blueprints in production manufacturing settings. |
Inspect, operate, and test completed products to verify functioning, machine capabilities, or conformance to customer specifications.
38CI 30–46 · exposure 38 · augmentation 63 · importance 4.5/5 · click for rater detail
Inspect, operate, and test completed products to verify functioning, machine capabilities, or conformance to customer specifications.
38| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Automated inspection has been adopted in high-volume manufacturing (automotive, electronics) but remains uneven across assembly sectors. Many smaller and job-shop environments rely heavily on human inspection; adoption is growing but not yet dominant in the broader engine/machine assembly sector. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/assembly is a moderate-to-slow adopter of AI-driven inspection compared to information sectors; sensor-based automated testing exists but broad AI-driven adoption for this specific task is still limited and uneven. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered vision and test systems significantly assist human inspectors by flagging anomalies, documenting measurements, and automating routine checks, allowing inspectors to focus on complex judgment calls and customer specification edge cases. This augmentation substantially raises inspector productivity and decision quality. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-enabled vision systems, predictive analytics, and automated data logging can meaningfully assist human inspectors in flagging anomalies and speeding data review, even though full judgment and physical operation remain human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Automated vision systems and sensors can inspect and test many machine parameters (dimensional accuracy, functional tests), but human judgment is often needed for subtle defects, customer specification nuances, and complex conformance decisions. End-to-end automation with 50% time savings is achievable on routine inspection components but not consistently across all verification scenarios. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical inspection, operation, and testing of assembled engines/machines requires manual handling, sensory judgment, and often specialized test rigs that current general-purpose AI cannot perform end-to-end without substantial robotic hardware integration. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Liability and regulatory frameworks sometimes require human sign-off on critical safety specifications and customer conformance, creating friction. Organizational inertia and the need for human judgment on atypical products add moderate barriers, though no hard legal licensing requirement exists for automated inspection itself. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement typically, but quality/safety liability for verifying machine conformance to specifications creates strong incentive for human sign-off, especially in regulated industries like automotive or aerospace engines. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While inspection automation infrastructure has high upfront capital costs, the per-unit cost for a complete inspection worker replacement—including handling edge cases, integration, calibration, and oversight—is often comparable to or exceeds hourly labor costs for many assembly operations, especially in small-to-medium production runs. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Specialized test automation equipment plus integration and maintenance costs are substantial capital investments; for many mid-volume assembly operations this is not clearly cheaper than skilled human testers, though at very high volume it can be justified. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed automated inspection systems (computer vision, test benches) exist in manufacturing and function reliably for structured tasks, but they have material limitations with edge cases, novel configurations, and require human sign-off for customer-critical specifications. Products are in production but typically augment rather than replace human inspectors. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated test stands and vision-based inspection systems exist in some factories, but full autonomous inspect-operate-test cycles for complex machines like engines remain narrow, custom-engineered solutions rather than widely deployed general products. |
Verify conformance of parts to stock lists or blueprints, using measuring instruments such as calipers, gauges, or micrometers.
37CI 25–49 · exposure 38 · augmentation 50 · importance 4.4/5 · click for rater detail
Verify conformance of parts to stock lists or blueprints, using measuring instruments such as calipers, gauges, or micrometers.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing adoption of AI inspection is occurring primarily in large aerospace and automotive suppliers; small and medium job shops and contract manufacturers lag significantly. Broad deployment remains limited due to the heterogeneous nature of parts, setups, and blueprints across different customers and product lines. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and assembly are historically slower adopters of AI-driven automation compared to information/professional services, though automated inspection is a known industrial application with steady but gradual uptake. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted measurement systems can help operators by automatically flagging out-of-spec dimensions or highlighting measurement regions on blueprints, reducing cognitive load and speeding verification tasks. However, augmentation is limited by the current lack of seamless integration with existing hand-tool workflows and the operator's need to make final judgments. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Digital calipers/gauges with data logging, and AI-assisted vision-based measurement tools can speed up verification and reduce transcription errors, meaningfully aiding human inspectors without fully replacing them. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI vision systems can identify and measure parts with some accuracy using image analysis, but the task requires high-precision conformance checking against blueprints and stock lists in varied physical environments. Setup and integration complexity, combined with the need for human verification of edge cases, means less than 50% net time savings at equal quality with current systems. |
| Task automatability | claude-sonnet-5 | 2/5 | Automated inspection with vision systems and coordinate measuring machines can check some part conformance, but the manual use of calipers/gauges/micrometers on varied parts in assembly settings still requires human dexterity and judgment for many cases today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Quality assurance in manufacturing carries high liability and error-cost asymmetry; defective parts reaching assembly lines or customers can result in costly recalls or safety failures. Regulatory frameworks (ISO, IATF, industry-specific standards) often require documented human inspection or sign-off, creating legal and compliance barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but quality/safety-critical components (e.g., engines) may require documented human sign-off or calibrated inspection procedures under quality standards like ISO/AS9100, creating moderate procedural friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI vision inspection systems require significant capital investment in cameras, lighting, hardware integration, and software maintenance, while the human task involves a worker with hand tools. For most small and medium assembly operations, the total cost of an AI system remains higher than paying a skilled operator to perform the verification. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Automated metrology equipment requires significant capital investment, fixturing, and integration engineering; for low-to-mid volume or varied part assembly work, this is often costlier than a trained assembler using hand tools. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Computer vision systems exist for parts inspection, but production deployments are still limited and often narrow in scope. Most real-world assembly verification relies on human operators with traditional measuring instruments; AI systems in production typically handle only constrained, well-lit scenarios or serve as an initial screening layer requiring human sign-off. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated dimensional inspection (vision systems, CMMs) is deployed in many manufacturing lines, but for the specific manual instrument-based verification described, robotic/automated solutions are narrower and often limited to high-volume, standardized parts. |
Maintain and lubricate parts or components.
33CI 30–35 · exposure 25 · augmentation 38 · importance 3.9/5 · click for rater detail
Maintain and lubricate parts or components.
33| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption is slow outside large manufacturers with highly standardized, high-volume production lines. Most engine and machine assembly remains in small-to-mid-sized shops with diverse equipment, limited capital for automation, and low digitization—classic laggard sectors. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/assembly sectors are historically slower adopters of AI-driven robotic automation for fine physical maintenance tasks compared to information-based industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted monitoring systems (condition-based maintenance alerts, wear prediction) can help technicians prioritize which parts to service, improving efficiency and reducing unplanned downtime. However, the physical act of lubrication still requires human presence and judgment. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can provide predictive maintenance scheduling or diagnostic alerts to guide when lubrication is needed, but it offers minimal direct assistance in performing the physical lubrication task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI-driven robots can perform some repetitive lubrication tasks in controlled factory settings, the task requires physical manipulation, sensory feedback (detecting wear, proper lubrication levels), and contextual judgment about which parts need service. Current systems cannot reliably handle the full end-to-end task across diverse machine types and conditions, so 50% time savings at equal quality is not yet achievable. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical manipulation task requiring hands-on manual work with parts, lubricants, and tools; current AI (software/LLMs) cannot perform the physical actions, though robotic automation exists for narrow, highly standardized cases.rb |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Some organizational and safety friction exists (equipment downtime, safety protocols), but there are no hard legal or licensing barriers preventing automation. Adoption is primarily constrained by technical maturity and ROI economics rather than regulation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but physical infrastructure, safety considerations, and the need for adaptable dexterity create moderate organizational and technical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic systems capable of maintenance tasks are expensive to deploy, program, and integrate; ongoing maintenance and oversight costs are high. For most assembly shops, the all-in cost remains higher than employing a human technician for this relatively low-wage task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic lubrication cells require significant capital investment, engineering, and maintenance, making them cost-competitive only in very high-volume, standardized settings; for typical variable assembly tasks human labor remains cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some industrial robots perform basic lubrication on fixed assembly lines, but these systems are narrow in scope and require significant setup and customization. No mature off-the-shelf product reliably performs maintenance and lubrication across varied machine assemblies without human oversight and adjustment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some fixed automated lubrication/maintenance systems exist in high-volume manufacturing, but general-purpose flexible maintenance and lubrication of varied components by robots is still limited to research or narrow deployments. |
Remove rough spots and smooth surfaces to fit, trim, or clean parts, using hand tools or power tools.
31CI 26–35 · exposure 20 · augmentation 25 · importance 4.2/5 · click for rater detail
Remove rough spots and smooth surfaces to fit, trim, or clean parts, using hand tools or power tools.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of full automation for this task is slow; most assemblers still rely on manual hand tools and power tools. While some advanced manufacturers use automated deburring, the heterogeneous nature of parts and the need for adaptive finishing keep most shops in a hybrid manual state. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/assembly is a slower-adopting sector for AI-driven automation compared to information/professional services; robotic finishing adoption is real but incremental and concentrated in large-scale production lines. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI offers minimal augmentation; power tools themselves are already well-established. AI vision systems could theoretically flag surface defects, but current systems are unreliable at detecting fit-related finish requirements, limiting practical assistance. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-guided vision systems or robotic assist tools can help identify defects or guide power tool operation in some settings, but general assistance for this hands-on task is limited compared to purely cognitive tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While power tools can automate some material removal, the task requires judgment about what constitutes 'fit' and determining when a surface is adequately smooth—decisions that demand tactile feedback and visual inspection that current AI systems cannot reliably replicate end-to-end. Manual intervention is needed for the precision fitting aspect. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical deburring/finishing task requiring tactile judgment and dexterous tool manipulation; current AI (software/robotics) cannot reliably replace this end-to-end without heavy custom robotic engineering, which is not off-the-shelf automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are few regulatory barriers to automation itself, but significant organizational friction exists: specialized tooling, machine setup, validation of fit quality, and the physical variability of parts create practical adoption resistance in typical assembly environments. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but physical variability of parts, need for tactile quality judgment, and capital investment in custom robotics create meaningful organizational and technical friction against substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current robotic solutions capable of any part of this work (CNC grinders, deburring machines) are capital-intensive and require setup costs that exceed the loaded wage of a skilled assembler for most production volumes. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic finishing systems require expensive fixturing, programming, and part-specific tooling, making them costly relative to a skilled assembler unless production volume is very high, so cost parity is achieved only in narrow high-volume cases. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system reliably performs this task autonomously in production. Robotic systems exist for narrow grinding/sanding tasks but lack the dexterity, environmental sensing, and adaptive decision-making needed for the full 'fit' and 'clean' workflow described here. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Robotic deburring and finishing cells exist in high-volume manufacturing for specific standardized parts, but they are narrow, part-specific setups rather than general-purpose products handling varied fit/trim/clean operations reliably. |
Set up and operate metalworking machines, such as milling or grinding machines, to shape or fabricate parts.
30CI 25–35 · exposure 25 · augmentation 50 · importance 4.0/5 · click for rater detail
Set up and operate metalworking machines, such as milling or grinding machines, to shape or fabricate parts.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Most manufacturing facilities still rely on skilled human operators and traditional CNC systems; adoption of fully autonomous adaptive machining is minimal outside large, high-volume automotive and aerospace operations, and even there, humans remain in the loop. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing has adopted CNC and robotics over decades, but this is distinct from generative AI adoption; physical production sectors show slower uptake of newer AI-driven automation compared to information sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist with tool-path optimization, predictive maintenance alerts, and quality-control flagging, meaningfully supporting operator productivity on parts of the task, though human judgment remains essential for setup decisions and responding to anomalies. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted CAM software, predictive maintenance, and machine vision inspection can improve setup accuracy and reduce errors, aiding but not replacing the operator's judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can control CNC machines for repetitive cuts, this task requires real-time perception of part geometry, tool wear, temperature monitoring, and adaptive decision-making in response to material variations that current systems struggle with reliably. Full end-to-end automation with 50% time savings at equal quality is not yet achievable with off-the-shelf systems. |
| Task automatability | claude-sonnet-5 | 2/5 | Setting up and operating milling/grinding machines requires physical manipulation, fixturing, and real-time sensory adjustment that current AI cannot perform end-to-end without robotic hardware and extensive integration.rn CNC automation exists but is a distinct technology from general AI performing the task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Setup and operation of metalworking machines often requires licensed or certified machinists in regulated manufacturing environments, and liability concerns (product defects, safety) create strong legal and organizational friction against full automation without human sign-off. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing barrier, but physical workspace safety, machine calibration expertise, and quality/tolerance liability create organizational friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Integration of vision systems, force sensors, and control logic for adaptive machining is capital-intensive and requires ongoing maintenance. The loaded cost per task-equivalent remains higher than employing a skilled machine operator, especially when factoring in setup and error recovery. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic/CNC automation requires significant capital investment in machinery, tooling, and integration, which is often costlier than skilled labor for small-to-medium batch or varied part production. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | CNC programming and operation are mature, but this task involves setup, manual adjustment, and responding to in-process problems. No deployed AI system reliably handles the full cycle—from interpreting part specifications to detecting misalignment, adjusting feeds/speeds, and quality verification—without human oversight. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | CNC and some adaptive machining systems exist in production, but 'setting up and operating' conventional milling/grinding machines still relies heavily on skilled human machinists for fixturing, tool changes, and quality checks. |
Fasten or install piping, fixtures, or wiring and electrical components to form assemblies or subassemblies, using hand tools, rivet guns, or welding equipment.
26CI 16–35 · exposure 17 · augmentation 25 · importance 4.3/5 · click for rater detail
Fasten or install piping, fixtures, or wiring and electrical components to form assemblies or subassemblies, using hand tools, rivet guns, or welding equipment.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Assembly automation has grown in automotive and electronics, but adoption remains concentrated in large-scale manufacturers with standardized processes; small and mid-size shops, custom assembly, and retrofit work continue relying on human labor due to setup costs and low-volume requirements. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing has moderate robotics adoption in high-volume standardized processes, but flexible, mixed-task assembly involving piping/wiring/fasteners sees slower uptake due to task variability and cost of reconfiguration. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI offers limited real-time assistance for this task; AR or robotic arms might guide part placement, but the core hand-tool operation, spatial problem-solving, and real-time adaptation to fit tolerances remain primarily human-driven with minimal productivity transformation from current AI tools. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Some augmentation exists via powered tools, torque-controlled fastening systems, and cobot assistance for holding/positioning, but no AI-driven decision support meaningfully transforms this manual task's productivity today. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While some aspects like fastening can be partially automated by robots in controlled manufacturing settings, the task requires dexterous hand tool operation, precise spatial judgment, and adaptive responses to component variations that current general-purpose AI systems cannot reliably execute end-to-end. Real-world assembly involves irregular part alignment, fit tolerance adjustments, and tool selection that exceed today's automation capability. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manual assembly task requiring dexterity, force application, and fine motor control that current AI systems (software-based) cannot perform; robotics for this exact variable task remain limited to structured, high-volume lines rather than general assembly. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Assembly work in many sectors operates under quality control and safety certifications requiring documented human sign-off, liability for defects rests with manufacturers, and assembly of safety-critical systems (automotive, aerospace, electrical) faces regulatory requirements for human oversight or validation of work quality. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for most assembly work, though certified welding tasks may require qualified welders; main barriers are practical (workspace variability, tooling changeover) rather than regulatory or liability-driven. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Industrial robot systems capable of assembly tasks require substantial capital investment, facility integration, and maintenance, making per-unit costs comparable to or exceeding trained human assemblers for variable, low-volume work. Cost advantage emerges only in high-volume repetitive scenarios not typical of the broader assembly workforce. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Industrial robotic welding/riveting cells can be cost-effective at very high volume, but for varied, lower-volume assembly work the capital cost of robotic tooling and integration typically exceeds human labor cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Specialized industrial robots exist for specific repetitive fastening tasks in automotive and appliance sectors, but they operate in highly controlled environments with predetermined components. No deployed general-purpose AI system reliably performs the full range of fastening, piping alignment, and electrical component installation described here across varied assembly contexts. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Robotic arms perform some fastening/welding in fixed automotive lines, but flexible piping/wiring/fixture installation across varied engine/machine assemblies is still mostly done by human assemblers; deployed general-purpose solutions are narrow and setup-heavy. |
Rework, repair, or replace damaged parts or assemblies.
24CI 16–31 · exposure 20 · augmentation 38 · importance 4.1/5 · click for rater detail
Rework, repair, or replace damaged parts or assemblies.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing is slowly digitizing, but repair and rework remain highly manual activities with low AI adoption in production. Most facilities still rely on experienced technicians to assess and repair damage on a case-by-case basis. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/assembly sectors show slow, uneven AI adoption for physical rework tasks compared to information-based industries, with automation focused on scripted, repetitive assembly rather than repair. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by identifying damage types from images, suggesting repair procedures, and highlighting critical areas, but the human assembler remains essential for diagnosis, execution, and quality control on this task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with defect detection or diagnostic guidance via computer vision, but offers limited help with the actual manual rework and repair execution. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Reworking, repairing, or replacing damaged parts requires real-time visual inspection, spatial reasoning, and fine motor manipulation in physical space. While AI can identify some damage from images, end-to-end execution of complex assembly repairs with 50% time savings at equal quality remains well beyond current capabilities. |
| Task automatability | claude-sonnet-5 | 2/5 | Diagnosing and physically reworking damaged mechanical assemblies requires fine motor manipulation, force feedback, and judgment about non-standard damage patterns that current robotics/AI cannot handle end-to-end.no off-the-shelf system approaches the 50% time-saving bar. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety-critical assembly work often requires human sign-off and accountability; quality assurance, liability for failure, and the need for human judgment on non-standard damage create strong organizational and regulatory friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but quality/safety liability for reworked engine parts and physical dexterity needs create moderate organizational friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of robotic systems, vision systems, and integration required to attempt autonomous repair substantially exceeds the loaded wage of a skilled assembler, especially given the high error costs in assembly work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Flexible robotic manipulation and vision systems capable of handling varied damage and repair scenarios are far more expensive to develop and deploy than employing a skilled assembler for this variable task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI systems reliably perform physical assembly repairs autonomously in production. Robotic systems exist for structured, repetitive assembly, but detecting damage, diagnosing root causes, and executing repairs on varied assemblies remain primarily human tasks with AI only in early monitoring roles. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously reworks or repairs damaged mechanical assemblies in production; robotic rework remains research-stage or limited to highly constrained, pre-programmed fixes. |
Set and verify parts clearances.
23CI 16–30 · exposure 20 · augmentation 50 · importance 4.4/5 · click for rater detail
Set and verify parts clearances.
23| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Most machine assembly remains in small to mid-sized manufacturing settings with limited automation investment. While some large automotive plants have upgraded inspection systems, clearance verification has not seen broad AI-driven adoption outside aerospace and premium automotive segments. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing assembly is a physically-oriented, moderate-digitization sector where AI adoption for precision mechanical tasks remains slow compared to information-based industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted measurement tools (computer vision for gap detection, automated alerts for out-of-spec clearances) can help assemblers work faster and catch defects more reliably. These systems augment rather than replace human judgment in setting and finalizing clearances. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-enabled measurement tools and vision systems can assist assemblers by flagging out-of-tolerance clearances or logging data, improving accuracy and speed while the human still performs the physical setting. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Setting and verifying parts clearances requires precise spatial reasoning, tactile feedback, and judgment about acceptable tolerances in physical assembly. While vision systems can measure some clearances, the task involves subjective judgment and adaptive adjustment that current off-the-shelf AI cannot reliably perform end-to-end without significant human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical manipulation of parts and precision measurement tools (feeler gauges, calipers) in a hands-on assembly context, which current AI cannot perform end-to-end without robotic hardware specifically engineered for this task.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Manufacturing quality and safety standards often require human sign-off on critical clearances, and liability for assembly defects creates a strong organizational preference for human verification. Regulatory requirements in many sectors mandate documented human inspection for safety-critical assemblies. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing barrier exists, but quality/safety liability and the need for tactile precision and adaptability to part variance create meaningful organizational and technical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Custom vision systems and gauging equipment for clearance verification are expensive to set up and maintain relative to the loaded hourly wage of assemblers performing spot checks. Integration into existing assembly lines adds non-trivial overhead, making replacement economically marginal. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Deploying robotic/AI systems capable of precision clearance verification requires expensive specialized hardware, sensors, and integration far exceeding the cost of a trained assembler performing the same task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Machine vision can detect and measure gaps in controlled settings, but deployed solutions for automated clearance verification in live assembly remain limited and error-prone. Most production environments still rely on human inspectors or semi-automated systems requiring substantial human validation of results. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No widely deployed product autonomously sets and verifies mechanical clearances on engines/machines in production assembly lines; this remains largely a manual or fixed-automation (not AI-driven) process. |
Position or align components for assembly, manually or using hoists.
21CI 7–35 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail
Position or align components for assembly, manually or using hoists.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Assembly automation is concentrated in large automotive, electronics, and appliance manufacturers with high volumes and standardized parts. Smaller job shops, custom assemblers, and medium-volume manufacturers (where many engine and machine assemblers work) show slow adoption of flexible automation due to the high variety of tasks and lower economies of scale. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing/machine assembly is a physical, moderately digitized sector where robotic adoption is real but slow and capital-intensive, especially for flexible non-repetitive tasks like this. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AR-guided positioning systems, real-time torque and alignment feedback tools, and digital assembly instructions can meaningfully assist assemblers in reducing errors and speeding up component positioning. These assist human workers but do not displace them, enhancing their productivity on precision and alignment steps. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-guided vision systems or hoist automation can assist with positioning cues or load calculations, but the core manual alignment work sees limited productivity transformation from current AI tools. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Positioning and aligning components requires fine spatial reasoning, dexterity, and real-time tactile feedback in three-dimensional space. While robotic systems exist in controlled environments, current AI and general-purpose robots struggle with unstructured variations in component geometry, surface conditions, and assembly tolerances without extensive task-specific programming or custom tooling. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring dexterity, spatial judgment, and force control that current AI systems cannot perform end-to-end; robotics for flexible, varied manual assembly with hoists remains largely research/pilot stage. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Assembly quality directly affects product safety, liability, and warranty claims; any automation must meet strict quality and traceability standards. Many assembly jobs in smaller shops and specialized manufacturing also lack the digital infrastructure, standardized specifications, and capital availability that would enable easy automation adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but safety regulations around heavy equipment/hoist operation and liability for defective assembly create some organizational and safety-compliance friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic systems capable of precision positioning and alignment typically cost tens of thousands to hundreds of thousands in capital plus integration labor. For lower-volume or mixed-component assembly, total cost of ownership often exceeds that of paying workers direct wages and benefits, especially when accounting for maintenance and setup time. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Robotic systems capable of this flexible manual task require expensive custom engineering, sensors, and integration, making them costlier than human labor for most variable assembly work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Specialized industrial robots perform similar tasks in highly controlled manufacturing settings with fixed part geometries and pre-engineered jigs. However, general-purpose AI systems cannot reliably perform this task in diverse assembly contexts; deployed solutions require extensive customization and do not work reliably across different component types or assembly instructions. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No widely deployed product autonomously positions and aligns heterogeneous engine components using hoists in production assembly lines; fixed automation exists for narrow repetitive cases but not general flexible assembly. |
Related occupations — Production
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.