Foundry Mold and Coremakers

51-4071.00
Median wage $48,110/yr12,790 employed (US)Rank #666 of 923 scored · top 72% by substitution

Make or form wax or sand cores or molds used in the production of metal castings in foundries.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution20
Exposure11
Augmentation20

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

13 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%12

panel mean rating 1.5/5 → substitution pressure 12/100

Technical feasibility todayw 20%10

panel mean rating 1.4/5 → substitution pressure 10/100

Cost vs. human wagew 15%13

panel mean rating 1.5/5 → substitution pressure 13/100

Adoption barriersw 20%inverted — strong barriers lower the score56

panel mean rating 2.8/5 (barrier strength) → substitution pressure 56/100

Sector adoption velocityw 10%9

panel mean rating 1.3/5 → substitution pressure 9/100

Task breakdown (13 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.

Sprinkle or spray parting agents onto patterns and mold sections to facilitate removal of patterns from molds.

30

CI 28–33 · exposure 25 · augmentation 13 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Foundry work is in a traditional, capital-constrained, low-digitization sector with many small and mid-sized shops. Automation adoption in foundries lags manufacturing and is concentrated only in large facilities with high-volume standardized production.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physical manufacturing sector with historically slow adoption of automation technologies, especially AI-specific solutions.
Augmentation potentialclaude-haiku-4-5-202510012/5AI vision could help flag coverage gaps or suggest optimal spray patterns, but the physical act itself is already direct and operator-controlled; augmentation gains are minimal compared to simple hands-on spraying.
Augmentation potentialclaude-sonnet-51/5AI offers essentially no meaningful assistance to a human performing this specific manual spraying task, as it is a physical, non-cognitive action.
Task automatabilityclaude-haiku-4-5-202510012/5While spraying/sprinkling is mechanically simple, this task requires spatial awareness, coverage judgment, and pattern recognition in a 3D foundry environment with variable geometries. Current robotics can perform repetitive spraying on uniform surfaces, but adapting to diverse mold patterns and achieving consistent coverage remains difficult without significant customization.
Task automatabilityclaude-sonnet-52/5This is a physical manipulation task requiring dexterity and sensor-based control of spraying equipment; while robotic automation exists in some foundries, it is not an AI-driven cognitive task and current general-purpose AI cannot perform the physical action itself.atable via specialized robotics rather than AI per se. ratable low.
Adoption barriersclaude-haiku-4-5-202510013/5This task has moderate barriers: it requires no formal licensing, but foundries are safety-sensitive environments with process controls and union presence in some facilities, creating organizational friction to rapid automation adoption.
Adoption barriersclaude-sonnet-52/5No licensing barriers, but physical workspace safety, equipment retrofitting costs, and the need for physical presence in the foundry create moderate organizational friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5A robotic spray system with vision guidance, integration into foundry workflows, and ongoing maintenance would cost significantly more than the loaded wage of a foundry mold worker, especially given the low-volume, high-variety nature of mold work.
Cost vs. human wageclaude-sonnet-52/5Robotic spraying systems require significant capital investment for equipment and integration, which is often more costly than manual labor especially in small-to-mid-size foundries, though could be cheaper at very high volume.
Technical feasibility todayclaude-haiku-4-5-202510012/5Industrial spray robots exist but are typically configured for high-volume, standardized applications (auto parts, uniform surfaces). Foundry mold work involves highly variable patterns, shapes, and sizes; deployed solutions lack the flexibility and vision-based adaptability needed for this task at production scale.
Technical feasibility todayclaude-sonnet-52/5Automated spray/parting agent systems exist in some industrial foundries as fixed automation, but these are engineered robotic/mechanical solutions, not AI products, and adoption is limited to larger, capital-intensive operations.

Operate ovens or furnaces to bake cores or to melt, skim, and flux metal.

26

CI 23–30 · exposure 25 · augmentation 38 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Foundry operations remain labor-intensive, geographically dispersed, and technology-resistant sectors with low digitization and capital constraints. Adoption of AI agents in this domain is negligible; most foundries rely on traditional manual and legacy automated systems rather than cutting-edge AI.
Sector adoption velocityclaude-sonnet-52/5Manufacturing/foundry sectors are traditionally slow adopters of AI and robotics, with capital-intensive retrofit needs and low digitization compared to information-sector industries.
Augmentation potentialclaude-haiku-4-5-202510012/5Sensor dashboards and predictive maintenance alerts can support foundry workers, but AI assistance on the core task (skimming, fluxing, metal assessment) is minimal. The task remains highly dependent on operator experience and judgment, with limited scope for meaningful AI-driven productivity gains.
Augmentation potentialclaude-sonnet-53/5AI-enabled sensors and process control systems can assist by monitoring temperature, timing, and metal composition, improving consistency and safety while a human remains in charge of physical operation.
Task automatabilityclaude-haiku-4-5-202510012/5While temperature monitoring and timing could be partially automated via sensor systems, the skilled judgment required for metal quality assessment, manual skimming, fluxing, and real-time furnace adjustment demand human expertise. Current AI lacks the sensorimotor capability and domain knowledge to perform the full sequence reliably without significant human oversight.
Task automatabilityclaude-sonnet-52/5This is a physical, hands-on manufacturing task involving equipment operation, material handling, and sensory judgment (visual/thermal cues for skimming and fluxing) that current AI cannot perform end-to-end without robotics.that AI cannot substitute for today.), only narrow sub-components like temperature monitoring can be automated.
Adoption barriersclaude-haiku-4-5-202510014/5Foundry work involves workplace safety regulations, worker qualification requirements, and liability concerns around molten metal handling. Regulatory frameworks and insurance protocols typically require trained, licensed personnel to operate or directly oversee furnace operation, creating substantial legal and organizational barriers to full automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement, but safety regulations around molten metal handling, high liability for defects/accidents, and physical workspace constraints create moderate barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Implementing furnace automation requires expensive specialized sensors, controls, and integration into existing foundry infrastructure. When amortized against foundry worker wages and the high cost of system failures (molten metal hazards, material loss), the all-in cost remains comparable to or higher than skilled human labor.
Cost vs. human wageclaude-sonnet-52/5Automating this requires industrial robotics, sensors, and control systems with significant capital investment, oversight, and maintenance, making it costlier than human labor in most small-to-mid-size foundries.
Technical feasibility todayclaude-haiku-4-5-202510012/5Industrial furnace control systems exist but are narrow supervisory tools (alarms, logging) rather than autonomous agents that skim, flux, and assess metal quality. No deployed AI system reliably performs the full task end-to-end; human operators remain essential in production environments.
Technical feasibility todayclaude-sonnet-52/5Some automated furnace control systems exist in modern foundries, but full autonomous operation of ovens/furnaces including skimming and fluxing molten metal is not a mature deployed product; most foundries still rely on skilled operators.

Sift and pack sand into mold sections, core boxes, and pattern contours, using hand or pneumatic ramming tools.

25

CI 15–35 · exposure 13 · augmentation 25 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundries are capital-intensive but labor-cost-conscious operations with slower digital transformation and automation adoption than professional services or finance. Pilot projects exist, but widespread production deployment of sand-packing automation remains limited.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, heavy-industrial physical sector where AI and robotic adoption for this specific task is minimal and slow-moving compared to information-based sectors.
Augmentation potentialclaude-haiku-4-5-202510012/5AI and sensors could assist workers by detecting sand density anomalies or guiding pneumatic tool application, but the core task—physically sifting and packing—is not easily augmented by information systems without substantial hardware integration. Marginal benefits accrue mainly through monitoring, not through productivity multiplication.
Augmentation potentialclaude-sonnet-52/5AI could assist indirectly through process monitoring, sensor-based quality control, or predictive maintenance in the foundry, but it offers little direct augmentation to the physical act of sifting and ramming sand.
Task automatabilityclaude-haiku-4-5-202510012/5While sifting and packing sand involves repetitive physical motions, the task requires real-time visual perception, dexterous manipulation, and adaptation to irregular mold contours and pattern shapes. Current robotics can handle structured packing in controlled environments, but sifting sand with hand or pneumatic tools and conforming to variable pattern geometries remains beyond practical end-to-end automation at equal quality.
Task automatabilityclaude-sonnet-51/5This is a physical manual task requiring hand or pneumatic ramming of sand into molds; current AI (software/LLMs) has no capability to perform physical manipulation, and general-purpose robotics has not achieved deployable competence for this specific foundry task.
Adoption barriersclaude-haiku-4-5-202510012/5Foundry work is not heavily regulated in terms of automation prohibitions, but occupational safety and union presence in some foundries create friction. Customer and organizational inertia around changing established molding practices also slows substitution.
Adoption barriersclaude-sonnet-52/5No licensing or legal requirement mandates a human perform this specific task, but physical workplace realities, capital costs of specialized automation, and the sensory/tactile judgment involved create moderate practical barriers to AI-driven substitution.
Cost vs. human wageclaude-haiku-4-5-202510012/5Foundry wages are modest, but deploying and maintaining specialized molding robots (including vision systems, pneumatic integration, and changeovers) remains expensive relative to the hourly labor cost. Integration and per-mold customization add overhead that does not yet undercut manual labor on typical jobs.
Cost vs. human wageclaude-sonnet-51/5There is no general AI system that performs this physical task, so AI cost per task-equivalent is effectively infinite or inapplicable; any automation would require expensive custom robotics/machinery, not cheaper than human labor.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some industrial robots perform basic sand packing in foundries, but deployed systems are limited to highly standardized mold sections and require significant setup per mold type. No mainstream product reliably handles the full task—sifting, adapting to contour variation, and quality inspection—across the diversity of cores and patterns in production foundries.
Technical feasibility todayclaude-sonnet-51/5No deployed AI or robotic product reliably performs sand sifting and packing into mold/core boxes in production foundries; this remains a manual or specialized fixed-automation process, not an AI-driven one.

Position cores into lower sections of molds, and reassemble molds for pouring.

24

CI 14–35 · exposure 13 · augmentation 25 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundries are traditionally capital-constrained, small to mid-size operations with low digitization. Automation adoption in foundry core and mold work has been slow and concentrated in large, high-volume facilities; most sectors show pilot interest rather than deep production deployment of this specific task.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, heavy manufacturing sector where robotic adoption is slow and uneven, especially in small-to-mid-size operations, with no evidence of rapid AI-driven displacement.
Augmentation potentialclaude-haiku-4-5-202510012/5AI vision can assist in mold defect detection or core positioning guidance, but the core assembly task itself—requiring hands-on spatial manipulation—offers limited augmentation potential; human workers benefit more from better tooling than AI assistance.
Augmentation potentialclaude-sonnet-52/5AI/vision systems could assist with quality inspection or guidance for core placement, but they offer limited assistance to the core physical positioning task itself.
Task automatabilityclaude-haiku-4-5-202510012/5Positioning cores and reassembling molds requires precise physical manipulation in a 3D environment with complex spatial reasoning and fine motor control. While AI vision systems can assess mold positions, current robotics struggle with the dexterity, adaptability, and real-time error correction needed for this highly variable manual task, achieving far less than 50% time savings at equal quality.
Task automatabilityclaude-sonnet-51/5This requires precise physical manipulation of heavy, variably-shaped cores and mold sections in a foundry environment, which current AI systems cannot perform end-to-end; robotic automation exists but is not general 'AI' automation in the software sense.'
Adoption barriersclaude-haiku-4-5-202510013/5Foundry work involves workplace safety regulations and liability for defective castings, creating some friction. However, there are no legal licensing requirements mandating a human sign-off, and safety barriers can be engineered into automation, leaving moderate rather than hard barriers to substitution.
Adoption barriersclaude-sonnet-52/5No licensing requirement, but physical workspace safety, precision tolerances, and the need for human dexterity/adaptability to mold variability creates real operational friction against pure automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Robotic systems capable of precision foundry work (custom grippers, vision integration, safety barriers) carry high capital and maintenance costs that exceed the loaded wage of foundry workers for most facilities, especially in small to mid-sized shops with variable molds.
Cost vs. human wageclaude-sonnet-52/5Custom robotic cells for core setting can be cost-effective in high-volume foundries, but require significant capital investment, engineering, and maintenance compared to a human coremaker's flexible labor cost.
Technical feasibility todayclaude-haiku-4-5-202510012/5No mainstream production systems reliably perform end-to-end mold core positioning and reassembly at scale. Robotic arms exist but require extensive task-specific programming and fail frequently on mold variants; deployed solutions are narrow and immature compared to human workers.
Technical feasibility todayclaude-sonnet-51/5No widely deployed AI product performs mold/core assembly; any automation here is specialized industrial robotics/fixed automation engineered per-line, not a generalizable AI product.

Tend machines that bond cope and drag together to form completed shell molds.

24

CI 14–35 · exposure 13 · augmentation 13 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundries are capital-constrained, geographically dispersed, and lag in digitization compared to automotive or electronics. Adoption of autonomous molding systems is slow and concentrated in large, well-capitalized facilities; most small and mid-tier foundries still rely on manual operation.
Sector adoption velocityclaude-sonnet-51/5Foundry/manufacturing work is a low-digitization, physical-labor sector with historically slow adoption of advanced automation and minimal AI-driven robotic deployment for this specific task.
Augmentation potentialclaude-haiku-4-5-202510012/5AI-assisted monitoring (computer vision for defect detection, predictive alerts) could aid operators, but the core task—physically tending the bonding machine—offers limited augmentation potential. Feedback systems might improve efficiency, but human judgment remains central and the task does not transform significantly with current AI assistance.
Augmentation potentialclaude-sonnet-51/5AI offers little direct assistance to a worker physically tending a bonding machine in real time, as this is a hands-on, low-cognitive-content operational task.
Task automatabilityclaude-haiku-4-5-202510012/5The task involves machine operation and monitoring that is highly repetitive and rule-based, but requires real-time physical interaction, sensory feedback (detecting misalignment, material defects), and occasional manual intervention. Current AI/robotic systems struggle with the dexterity, force feedback, and adaptive problem-solving needed for reliable end-to-end operation at human productivity levels.
Task automatabilityclaude-sonnet-51/5This is a physical machine-tending task requiring manual handling, positioning, and monitoring of foundry equipment on a factory floor, which current AI systems cannot perform end-to-end without robotics.dvancements far beyond typical software AI.dfd.
Adoption barriersclaude-haiku-4-5-202510013/5Physical task in a manufacturing environment with some regulatory safety requirements (OSHA, equipment standards), and customer/quality expectations around mold integrity create moderate friction. However, no licensing requirement or hard legal barrier mandates human oversight, and unions vary in strength across foundries.
Adoption barriersclaude-sonnet-52/5No licensing requirement exists for this manual task, but physical workplace safety standards, capital investment in specialized equipment, and reliability requirements create moderate organizational friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5Specialized robotic systems for foundry molding remain capital-intensive and require significant integration costs, safety systems, and maintenance. The loaded wage for foundry operators is modest, making the total cost of ownership for reliable automation comparable to or higher than sustained human labor in many foundries.
Cost vs. human wageclaude-sonnet-52/5Existing automation solutions in foundries rely on hardwired industrial machinery/PLCs rather than AI, and outfitting with flexible AI-driven robotics would be costly relative to a machine operator's wage.
Technical feasibility todayclaude-haiku-4-5-202510012/5While industrial automation and robotic arms exist for foundry work, consistent production-scale deployment of systems that autonomously tend bonding machines without human oversight remains limited. Most commercial solutions still require human operators for setup, quality checks, and fault recovery.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product tends foundry bonding machines autonomously; this remains a physical robotics/automation challenge, not a software AI capability, and is research-stage at best for full autonomy.

Position patterns inside mold sections, and clamp sections together.

20

CI 10–30 · exposure 8 · augmentation 13 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundry operations remain a traditional, labor-intensive sector with slower technology adoption than knowledge work or even other manufacturing. Most foundries still rely on manual positioning and clamping; AI-driven automation is rare in production and concentrated in large, capital-rich facilities.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a physically intensive, lower-digitization manufacturing sector where AI/robotic adoption for such specific manual tasks remains slow and limited to large-scale high-volume operations using fixed automation, not AI agents.
Augmentation potentialclaude-haiku-4-5-202510012/5Vision-assisted guidance or robotic arms under human control could provide some productivity gains, but the tight coupling of physical precision and judgment in pattern alignment limits meaningful augmentation. The task is already manual, and assistance tools are not yet standard.
Augmentation potentialclaude-sonnet-51/5Current AI offers no meaningful real-time assistance to a worker physically positioning and clamping mold sections; this is a hands-on physical task outside the scope of typical AI augmentation tools.
Task automatabilityclaude-haiku-4-5-202510012/5This task involves physical manipulation in a constrained industrial environment. While some positioning steps could theoretically be guided by vision systems, the clamping operation requires precise force control and real-time tactile feedback that current robotics struggles with reliably, and the task remains largely manual on factory floors today.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring dexterity, force application, and precise alignment of heavy pattern pieces inside mold boxes—current AI systems (software/LLMs) cannot perform this, and robotics for this specific unstructured task are not off-the-shelf solutions.
Adoption barriersclaude-haiku-4-5-202510013/5Foundry work carries occupational safety requirements and operator licensing in some jurisdictions, creating moderate friction. Liability concerns around clamping errors (worker safety) and equipment damage also slow automation adoption, though no absolute legal prohibition exists.
Adoption barriersclaude-sonnet-52/5No licensing requirement for this specific manual step, but industrial safety standards, custom tooling needs, and physical workspace constraints create moderate friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Industrial robots capable of handling mold clamping operations are capital-intensive (six figures to millions), with high integration costs for each foundry layout. The loaded cost per task cycle likely exceeds that of skilled foundry workers for most small-to-medium foundries, offsetting labor savings.
Cost vs. human wageclaude-sonnet-51/5There is no generally available AI system performing this physical task, so any comparison would require custom robotics with high capital and integration costs likely exceeding human labor costs for this task alone.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed commercial product reliably performs this full task end-to-end in foundry settings. Specialized industrial robots exist for specific mold operations, but general pattern positioning and clamping with the variability of foundry work remains in pilot or custom-engineering phases, not production at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed AI/robotic product reliably performs this specific foundry task in production; while some automated foundry lines exist for high-volume standardized molding, general pattern positioning and clamping across varied mold sections remains manual or fixed-automation, not AI-driven.

Lift upper mold sections from lower sections, and remove molded patterns.

20

CI 5–35 · exposure 13 · augmentation 25 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundries, particularly smaller and mid-sized operations, are traditionally laggard in automation adoption due to capital constraints, custom mold variability, and the physical, low-digitization nature of the work. Pilot projects exist but widespread production deployment remains limited.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physically intensive manufacturing sector with historically slow adoption of advanced automation compared to information-based industries.
Augmentation potentialclaude-haiku-4-5-202510012/5AI and automation offer minimal augmentation to foundry workers performing this task; vision systems could assist in mold alignment detection, but human operators remain the primary agents. The task is inherently physical and requires direct human judgment about pattern integrity and safe removal.
Augmentation potentialclaude-sonnet-52/5AI offers limited assistance to this purely physical task, though vision systems or robotic arms could someday provide minor support in mold inspection or handling guidance.
Task automatabilityclaude-haiku-4-5-202510012/5While robotic systems can lift and manipulate mold sections in controlled factory settings, the task requires perception of mold alignment, pattern fragility assessment, and adaptive force control that current AI-driven robotics struggles with reliably. End-to-end automation with 50% time savings at equal quality is not yet demonstrated at scale in production foundries.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task involving lifting heavy mold sections and separating molded patterns, requiring dexterity and force in a foundry environment that current AI systems cannot perform end-to-end without embodied robotics infrastructure not in general deployment.
Adoption barriersclaude-haiku-4-5-202510014/5Physical safety regulations, OSHA compliance, and liability concerns around equipment damage and worker injury create significant organizational and regulatory friction. Foundries are risk-averse about automating mold handling due to high-value mold damage costs and the legal requirement for proper equipment operation oversight.
Adoption barriersclaude-sonnet-52/5No licensing requirement exists for this manual task, but physical workspace constraints, variable mold sizes/weights, and safety considerations create moderate organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Robotic systems capable of safe mold handling and pattern removal require substantial capital investment, integration, and ongoing maintenance, while a foundry worker's loaded wage is relatively low. The all-in cost of automation remains higher than human labor for this specific task in most foundries.
Cost vs. human wageclaude-sonnet-51/5Physical robotic automation for foundry mold handling requires expensive custom engineering, integration, and maintenance that typically exceeds the cost of human labor for this task at current adoption levels.
Technical feasibility todayclaude-haiku-4-5-202510012/5Specialized industrial robots exist for foundry automation, but this specific task—lifting upper mold sections and removing patterns without damage—involves significant variation in mold geometry, weight distribution, and pattern integrity that deployed systems handle inconsistently. Production deployment is limited and material error rates remain high.
Technical feasibility todayclaude-sonnet-51/5No deployed products perform this specific foundry mold-handling task reliably in production; industrial robotics for foundry work exists but is narrow, custom-engineered, and not a general-purpose AI product.

Cut spouts, runner holes, and sprue holes into molds.

20

CI 10–30 · exposure 13 · augmentation 25 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundries are predominantly small to mid-sized, capital-constrained manufacturers with low digital integration. Adoption of advanced automation and AI is slow; most foundries still rely on manual and semi-automated processes, reflecting the laggard character of the sector.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physical manufacturing sector with slow automation adoption outside of large-scale die casting operations using dedicated machinery, not AI systems.
Augmentation potentialclaude-haiku-4-5-202510012/5AI-assisted vision systems could help inspect mold geometry or guide operators to optimal cutting paths, but the core manual skill of executing precise cuts remains largely unaugmented by current AI; assistance is limited to planning and inspection phases.
Augmentation potentialclaude-sonnet-52/5AI could assist with CAD-based mold design or process planning for spout/runner placement, but offers little direct assistance to the physical cutting action itself.
Task automatabilityclaude-haiku-4-5-202510012/5Cutting precise holes and spouts into molds requires fine motor control, 3D spatial reasoning, and real-time adaptation to material variation. While some cutting operations can be partially automated with CNC, the task involves context-dependent decisions about placement and depth that would require significant setup and human oversight to achieve equal quality.
Task automatabilityclaude-sonnet-51/5This is a precise physical manual task requiring hand tools or manual machining on molds; current AI systems have no general capability to perform this physical cutting operation end-to-end.
Adoption barriersclaude-haiku-4-5-202510013/5Foundry work involves skilled craft knowledge and regulatory safety requirements around molten metal and equipment operation. Operators must be trained and licensed; liability for mold defects rests with human workers and quality checks, creating moderate organizational and legal friction against full automation.
Adoption barriersclaude-sonnet-53/5No licensing barrier exists, but physical process control, safety standards, and quality/tolerance requirements in foundry work create meaningful organizational and technical friction against pure AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510012/5Foundry equipment and mold-cutting CNC systems are capital-intensive; integration with AI vision and planning systems adds cost. For small-batch and custom work typical in foundries, the all-in cost of automation often exceeds the wage cost of a skilled mold maker.
Cost vs. human wageclaude-sonnet-51/5Robotic or automated cutting systems for this niche task require significant capital investment, custom engineering, and maintenance, making them more expensive per unit than skilled manual labor in most foundry operations.
Technical feasibility todayclaude-haiku-4-5-202510012/5CNC and robotic systems exist for mold cutting in limited, high-volume production contexts, but current AI-integrated systems do not reliably handle the variability and precision demands of custom mold work at production scale without substantial human intervention.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs this specific foundry mold-cutting task; any automation here would rely on dedicated CNC/robotic tooling, not general AI systems, and such deployment is rare and narrow.

Clean and smooth molds, cores, and core boxes, and repair surface imperfections.

18

CI 10–26 · exposure 8 · augmentation 13 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundry work is in a capital-intensive, traditionally low-digitization sector with high physical-world variability; while some larger facilities pilot automation, adoption remains limited and slow compared to information or professional services sectors.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physical manufacturing sector with minimal AI/robotics adoption for this specific hands-on finishing task.
Augmentation potentialclaude-haiku-4-5-202510012/5Current AI offers limited assistance on this task; robotic arm guidance or vision-aided defect detection might help identify problem areas, but the core tactile and judgment work remains primarily human-driven with minimal productivity lift.
Augmentation potentialclaude-sonnet-51/5Current AI offers essentially no assistance to a worker physically cleaning, smoothing, or repairing mold surfaces; this is not a cognitive or data-driven task.
Task automatabilityclaude-haiku-4-5-202510012/5Manual cleaning and smoothing of foundry molds requires fine tactile feedback, spatial judgment, and judgment about surface quality that current automation struggles with at scale. While some abrasive or chemical cleaning steps might be partially automatable, the repair of surface imperfections requires human-level perception and adaptation that falls far short of 50% time savings with equal quality today.
Task automatabilityclaude-sonnet-51/5This is a physical manual task requiring hand tools and tactile inspection to clean, smooth, and repair molds/cores; no off-the-shelf AI system can perform this physical labor.'
Adoption barriersclaude-haiku-4-5-202510013/5Workplace safety regulations and quality standards require oversight of surface finishing, and foundries often prefer human judgment for critical mold preparation; however, no licensing requirement specifically protects the human performer, creating moderate but not hard barriers to substitution.
Adoption barriersclaude-sonnet-52/5No licensing barrier exists, but the physical nature of the work (dexterity, tactile feedback, variable defect types) creates strong practical barriers to automation absent specialized robotics investment.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized robotic systems capable of mold handling and surface finishing remain expensive relative to skilled foundry workers, and integration costs for customizing to varying mold geometries make the all-in cost per task substantially higher than human labor.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute for this physical labor, so any 'AI cost' would require robotics far beyond current deployed capability, making it more expensive or simply infeasible.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed products reliably perform the full task of cleaning, smoothing, and repairing surface imperfections on foundry molds in production. Robotic systems exist for limited cleaning applications but lack the dexterity and judgment needed for quality-critical surface repair.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical mold cleaning, smoothing, or surface repair; this remains a purely manual foundry task.

Pour molten metal into molds, manually or with crane ladles.

18

CI 5–30 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Foundry manufacturing is a traditional, lower-digitization sector with many small to mid-sized shops. While large foundries have adopted some robotic pouring, overall sector adoption remains slow and uneven. Pilot projects exist but broad production deployment is uncommon relative to information-sector automation.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, heavy manufacturing sector with slow technology adoption cycles, dominated by mechanical/robotic automation investments rather than AI agents.
Augmentation potentialclaude-haiku-4-5-202510013/5AI and sensor systems can assist foundry workers by monitoring mold temperature, predicting defects, optimizing pour timing, and alerting workers to deviations—raising safety and quality. However, the human remains essential for real-time judgment and physical control, limiting augmentation to partial productivity gains on specific subtasks.
Augmentation potentialclaude-sonnet-52/5AI could assist with monitoring pour timing, temperature sensors, and predictive quality control, but does not meaningfully augment the physical pouring act itself.
Task automatabilityclaude-haiku-4-5-202510012/5Pouring molten metal into molds involves physical manipulation in a hazardous environment with real-time thermal and spatial judgments. While robotic systems exist for some casting operations, they require substantial customization per foundry setup, mold geometry, and metal properties. Current AI cannot reliably orchestrate the full task—positioning, temperature monitoring, pour rate adjustment, and defect detection—end-to-end with 50% time savings at equal quality across diverse real-world scenarios.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring dexterous handling of molten metal via manual pouring or crane ladle operation; no off-the-shelf AI system can perform this physical action today.
Adoption barriersclaude-haiku-4-5-202510014/5Worker safety regulations, occupational health standards, and liability for burns or injuries create strong barriers to full automation. Many foundries face union agreements protecting workers, and regulatory oversight of foundry operations (OSHA, environmental) discourages unproven automation in hazardous tasks. Customer expectations and craft tradition also favor human oversight of casting quality.
Adoption barriersclaude-sonnet-53/5Extreme safety hazards (molten metal, high heat, risk of severe injury) create strong operational and liability barriers to introducing new automation, though this is more a physical/safety barrier than a regulatory-licensing one.
Cost vs. human wageclaude-haiku-4-5-202510012/5Specialized robotic pouring systems are capital-intensive (hundreds of thousands to millions of dollars) with high integration costs, often exceeding the loaded wage of foundry workers over medium timescales, especially for smaller operations or job-shop foundries with mixed production.
Cost vs. human wageclaude-sonnet-51/5There is no general AI system that can substitute for this physical task, so any comparison would require dedicated robotic foundry equipment which has high capital costs versus labor, not an AI inference cost comparison.
Technical feasibility todayclaude-haiku-4-5-202510012/5Industrial robots can pour metal in structured, high-volume settings with fixed mold geometries and standardized procedures. However, deployment is limited to large-scale facilities with significant capital investment; most foundries still rely on manual pouring. Reliability remains material for variable mold types, metal compositions, and thermal conditions, and no mass-market AI product performs this task autonomously in general foundry settings.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs molten metal pouring; this remains a human or specialized industrial automation (not general AI) task, and such automation is fixed hardware, not adaptable AI systems.

Rotate sweep boards around spindles to make symmetrical molds for convex impressions.

15

CI 15–15 · exposure 0 · augmentation 0 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5The foundry sector is a laggard in automation adoption relative to information and finance sectors; mold-making remains largely manual and concentrated in smaller, less digitized operations with limited capital for robotics.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physical manufacturing trade with minimal AI adoption; automation here historically comes from mechanical/robotic tooling rather than AI systems, and uptake is slow.
Augmentation potentialclaude-haiku-4-5-202510011/5This physical craft task offers minimal opportunity for AI augmentation, as the skilled worker directly manipulates materials and tools; AI cannot meaningfully assist in the core rotation and symmetry judgment without physical presence.
Augmentation potentialclaude-sonnet-51/5Current AI offers essentially no direct assistance to the physical act of rotating sweep boards around spindles; any augmentation would be indirect (e.g., design software) rather than to this specific manual task.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires precise physical manipulation of sweep boards around spindles in a three-dimensional space to create symmetrical molds. Current AI systems cannot operate physical machinery or perform fine motor control in unstructured foundry environments.
Task automatabilityclaude-sonnet-51/5This is a precise physical manipulation task requiring manual dexterity and real-time tactile feedback with sand/mold material; no off-the-shelf AI system can perform this physical operation today.
Adoption barriersclaude-haiku-4-5-202510012/5While there are no strict licensing barriers preventing automation, the physical and craft nature of the work, combined with the small scale and fragmentation of foundry operations, creates organizational friction against adoption.
Adoption barriersclaude-sonnet-52/5No licensing or legal requirement mandates a human perform this specific task, but physical workspace integration, capital equipment costs, and craft-skill specificity create moderate practical friction against substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized robotic arms capable of rotating sweep boards with the precision required for symmetrical convex impressions remain far more expensive than a skilled foundry worker's loaded wage, including integration and maintenance costs.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI-based substitute; any automation would require expensive custom robotics/mechanical fixtures, not standard AI inference, making cost comparison to a human coremaker unfavorable at current maturity.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI products perform physical mold-making operations. This is a hands-on manufacturing task requiring embodied robotics in a specialized, legacy-heavy industry where such deployment has not materialized.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs sweep-board mold making; this remains a specialized manual foundry craft skill, though robotic automation exists it is not general AI-driven and is narrow, custom industrial equipment rather than AI.

Move and position workpieces, such as mold sections, patterns, and bottom boards, using cranes, or signal others to move workpieces.

12

CI 5–19 · exposure 8 · augmentation 25 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Foundry operations are typically in small to mid-sized firms with lower capital investment in automation. Adoption of advanced robotic systems has been slow relative to other manufacturing sectors, with much work still performed manually or with basic mechanical assistance.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physical manufacturing sector with historically slow adoption of advanced automation, especially for bespoke, variable-shaped workpieces.
Augmentation potentialclaude-haiku-4-5-202510012/5While powered lifting equipment (hoists, chain hoists) provide some assistive benefit, AI systems offer minimal augmentation potential for the core task of positioning workpieces, as the work remains fundamentally manual and requires human spatial judgment.
Augmentation potentialclaude-sonnet-52/5AI-based vision or scheduling systems could assist in planning crane movements or signaling logistics, but the core physical positioning task sees minimal current AI augmentation.
Task automatabilityclaude-haiku-4-5-202510011/5This task involves physical manipulation of heavy industrial objects in a foundry environment, requiring real-time spatial coordination, safety awareness, and interaction with heavy machinery. Current AI systems cannot physically operate cranes or move objects, and no autonomous robotic systems are widely deployed for this specific foundry context.
Task automatabilityclaude-sonnet-52/5This requires physical manipulation of heavy mold sections via crane operation, which is a physical task not addressable by current AI software; robotic automation exists but is not 'AI' in the LLM/agent sense and requires heavy capital investment specific to each foundry layout.'
Adoption barriersclaude-haiku-4-5-202510014/5Significant barriers exist: foundry work often involves tight spaces, variable mold configurations, and safety-critical positioning that requires real-time human judgment. Additionally, workplace safety regulations and liability concerns around heavy machinery automation in foundries create substantial friction.
Adoption barriersclaude-sonnet-53/5No licensing requirement, but heavy machinery safety regulations, liability for dropped/misplaced heavy workpieces, and physical workspace constraints create meaningful friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The capital cost of deploying robotic crane systems or autonomous material handlers in foundries would far exceed the loaded wage of foundry workers, particularly when considering the specialized setup required for variable workpieces and positions.
Cost vs. human wageclaude-sonnet-51/5Deploying robotic crane systems with perception and control for irregular mold sections requires substantial capital and engineering investment, likely exceeding the cost of a human operator for most foundries.
Technical feasibility todayclaude-haiku-4-5-202510011/5While some industrial robots and autonomous systems exist, they are not deployed at scale for the specific task of moving and positioning foundry mold sections and patterns in typical foundry operations. This remains a human-performed task in production.
Technical feasibility todayclaude-sonnet-51/5No generally available AI product performs crane-based positioning of foundry mold sections; this remains a specialized industrial robotics/automation problem, largely research or custom-engineered rather than off-the-shelf deployed AI.

Form and assemble slab cores around patterns, and position wire in mold sections to reinforce molds, using hand tools and glue.

10

CI 5–15 · exposure 0 · augmentation 13 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Foundries are typically small-to-medium manufacturers with low digital maturity, high capital constraints, and entrenched manual labor practices. Adoption of advanced automation in core-making remains minimal, with most foundries continuing traditional hand methods.
Sector adoption velocityclaude-sonnet-51/5Foundry work is a low-digitization, physically intensive manufacturing sector with minimal AI/robotics adoption for tasks like manual mold and core assembly.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could marginally assist through design visualization or quality control suggestion, but the core task—physical formation and assembly of cores with hand tools—offers minimal augmentation opportunity since the human must perform the manual work regardless.
Augmentation potentialclaude-sonnet-51/5Current AI tools offer essentially no assistance for the physical, tactile work of shaping cores and positioning reinforcing wire by hand.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires precise 3D spatial manipulation, physical dexterity, and adaptation to variable foundry patterns—capabilities that current AI systems cannot perform end-to-end in unstructured factory environments. Hand-tool operation, glue application, and wire positioning around custom patterns remain firmly in the domain of skilled manual labor.
Task automatabilityclaude-sonnet-51/5This is a hands-on physical manipulation task requiring dexterity, tactile feedback, and adaptation to variable materials (sand, wire, patterns) that current AI and general-purpose robotics cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Foundry work is a skilled trade with apprenticeship and certification expectations; liability for defective molds carries high cost (failed castings, safety hazards). Additionally, the physical, on-site nature of the work and the need for real-time quality judgment create both regulatory and practical friction against full automation.
Adoption barriersclaude-sonnet-52/5No licensing requirement exists, but physical workspace constraints, variable pattern geometries, and lack of standardized robotic tooling create significant practical friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current robotic and AI solutions capable of handling flexible mold assembly would require significant capital investment, custom integration, and ongoing maintenance—all substantially more expensive than employing skilled foundry workers at standard wages.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI/robotic substitute performing this physical assembly task, so any hypothetical automation would require costly custom robotics far exceeding current human labor costs.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI systems or robotics products reliably perform the full assembly of slab cores with hand tools and glue in production foundries today. While research into robotic assembly exists, the task's variability, tactile feedback requirements, and need for real-time adaptation make production-scale feasibility absent.
Technical feasibility todayclaude-sonnet-51/5No deployed commercial product performs slab core assembly, wire reinforcement, or mold construction in foundries today; this remains a manual craft trade task.

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.