Nature Architects Inc. develops a software platform that deeply integrates geometry processing, numerical analysis, and optimization, while also providing design solutions enabled by the platform through its Generative Engineering™ approach. Covering a wide range of manufacturing contexts, from mass production to advanced manufacturing methods, the company dramatically accelerates design exploration and creates new designs beyond existing design paradigms.

This presentation will introduce case studies in which design concepts were generated under complex, large-scale physical phenomena and geometric constraints. Examples include the rapid exploration and creation of new design concepts for electric vehicle bodies, innovative garment design using origami-based techniques, and the design of highly efficient heat exchange structures enabled by additive manufacturing. The presentation　will also outline future directions for Generative Engineering™

Director and CTO of Nature Architects Inc.
After graduating from the Faculty of Engineering at the University of Tokyo, he entered the University of Tokyo’s Graduate School of Interdisciplinary Information Studies. While pursuing research in design engineering and working as an engineer at a product design studio, he joined Nature Architects as a founding member. He received the Software Japan Award in 2024

Computational design and additive manufacturing are expanding the possibilities for the aerospace industry, however, translating increasingly sophisticated geometries into flightworthy hardware continues to present significant qualification and industrialization challenges in the commercial and military landscapes. Three engineers from Honeywell Aerospace will share firsthand perspectives from their experience with computational design tools, metal additive manufacturing production and part qualification, and advanced technology research and product development, touching on real-world barriers that constrain the adoption of CDFAM toolsets and technologies in the aerospace industry. This conversation will highlight emerging opportunities where the CDFAM community can deliver innovative approaches, complementary capabilities, advanced toolsets, and collaborative pathways to support companies like Honeywell Aerospace with accelerating the integration state of the art design and manufacturing approaches into aerospace product development, production, and lifecycle management.

Scott Flower is an Advanced Additive Manufacturing Design Engineer at Honeywell Aerospace, where he is leading the internal adoption of implicit and generative design tools to deliver design solutions for next generation aerospace systems. Widely recognized as resident expert in advanced design workflows, Scott focuses on overcoming traditional CAD limitations to optimize and scale complex hardware for flight-certified industrial applications. His engineering approach bridges the gap between advanced computational design and digital fabrication methods such as Laser Power Bed Fusion, and he is actively investigating how to integrate artificial intelligence into aerospace product development cycles. Scott’s technical innovations include co-inventing patented manufacturing methods for Honeywell, such as advanced systems and methods for forming directionally solidified and single-crystal components to improve material performance in high-stress aerospace environments.

Curtis Swift serves as a Senior Advanced Additive Manufacturing Engineer and key manufacturing floor leader at Honeywell Aerospace’s Phoenix Additive Manufacturing Center, where he has been instrumental to the establishment of Honeywell’s specialized metal and vat-based ceramic 3D printing workflows for commercial aerospace and defense industry. His expertise in overcoming the technical, organizational, and cultural bottlenecks that typically slow down the industrial adoption of emerging additive technologies has enabled Honeywell Aerospace to reduce development timelines and additive manufacturing production risks, driving significant return on investment through the production of aerospace parts.

Ifeanyi J. Umunna is a Senior Research & Development Scientist and Technical Program Leader at Honeywell Aerospace, where he works on the maturation of advanced technologies for next generation aerospace systems. His work spans computation design, additive manufacturing, thermal management, advanced materials, and technology transition for Honeywell’s Engines & Power Systems and Materials and Mechanical Systems divisions, with focus of developing emerging technical concepts into valuable, manufacturable, and qualifiable aerospace products. He is a recipient of Honeywell’s Innovation and Growth award and has been invited to speak at the America Makes Technical Exchange, The Vertical Flight Society Technical Forum, and to the Honeywell Innovation Summit. Locally, in Phoenix, Arizona, Ifeanyi leads community STEM initiatives, mentoring students and faculty in local school districts, and supports community workforce development initiatives related to the growing sustainability and STEM economy.

What could the next generation of engineering software feel like, and what becomes possible when the interface is no longer the constraint?

The tools we use to create are changing. AI is making it possible to move beyond traditional software interfaces, but the opportunity isn’t simply to add AI to the interfaces we already have. It’s to rethink the interaction layer itself. Explore what happens when the interface becomes more fluid, more exploratory, and more directly connected to the human act of engineering

Laurence Cook is Co-Founder and Chief Product Officer at Generative Engineering.
After academic work at MIT, Stanford, and Cambridge, Laurence brought his computational engineering to the commercial industry and is now leading product at Generative Engineering.

Beatrice has built products at both ends of engineering, from electric vehicles to the infrastructure behind large-scale virtual worlds. Now Head of Product at Generative Engineering, she brings the physical and the digital together.

Artificial intelligence is poised to automate a large share of design engineering work, yet the technology that excites the commercial world poses a fundamental problem for high-consequence industries. Generative AI is probabilistic by nature. It produces plausible answers, not provably correct ones. In sectors where a single structural failure can ground a fleet, halt a production line, or cost lives, plausibility is not enough. The question is no longer whether AI will transform engineering, but whether we can trust it when failure is not an option.

This talk presents a different path. Cognitive Design Systems is a design exploration platform for mechanical and thermo-mechanical component design. Rather than embedding opaque AI inside traditional CAD software, we bring proven engineering workflows to the AI. Deterministic solvers for topology optimization, finite element analysis, manufacturing-driven design, and cost and carbon assessment produce repeatable, auditable, physically grounded results. A conversational AI layer orchestrates these solvers, interpreting intent and chaining tasks, while the underlying engineering computation remains fully deterministic and traceable. Engineers gain dramatic speed without surrendering verifiability or control.

This is not theoretical. Our approach is shaped by work with demanding industrial leaders including Safran, Thales, MBDA, Toyota, Tetra Pak, and Logitech, spanning aerospace, automotive, defense, and industrial machinery. These are organizations where engineering rigor and certification are non-negotiable.

The implications reach across every engineering sector. As manufacturers face mounting pressure to lightweight structures, accelerate certification, reduce cost and carbon, and modernize their industrial base, the ability to design qualified components faster, with full auditability, becomes a decisive advantage. Trustworthy AI is not a constraint on innovation. It is the precondition for deploying AI in the systems the world depends on. Attendees from industry and policy alike will leave with a clearer view of what responsible, deployable AI for high-consequence engineering actually looks like. 

Engineering never stands still. Every new product builds on last year’s designs and models, but staying competitive requires constant adaptation to solve the problems of tomorrow.

AI models recombine the past. They work well where past knowledge exists, but tomorrow’s market requirements go beyond what any model has seen before. Push requirements into that territory and they stop reasoning: they extrapolate into what they don’t understand, leaving engineers with no real confidence in the result.

The answer lies in treating engineering AI as an OS, not a tool: infrastructure that keeps models accurate and current as design requirements evolve, that learns from each new project rather than resetting, and that keeps proprietary knowledge where it belongs. Drawing on real industrial deployments, this talk explores what that infrastructure looks like in practice and what changes once engineering intelligence becomes part of how products get built.

Anna Kiener is the Lead Forward Deployed Engineer for APAC at Neural Concept, where she delivers Engineering Intelligence solutions for customers. She received her Ph.D. from the German Aerospace Center (DLR) and the Technical University Braunschweig on Data-Driven Corrections of Low-Fidelity CFD Simulations. This research naturally motivated her to bring AI-driven simulation methods from academia into real-world industry applications.

What if material fabrication could shift from assembly to growth—and be directed with precision through external fields?

This work introduces magnetotropic plants: genetically engineered organisms in which gravity-sensing organelles (statoliths) are rendered magnetically responsive. By replacing gravitational cues with externally applied magnetic fields, plant growth direction can be actively controlled in real time. This enables programmable morphogenesis, where biological growth becomes a steerable process rather than a fixed outcome of genetics and environment.
The presentation will outline the biological mechanism, the experimental framework, and the implications of this approach for material production.
Magnetic fields act as an invisible, non-contact control layer, allowing spatial and temporal guidance of growth without mechanical intervention.
Beyond applications in microgravity environments such as space, this work suggests a broader shift in how we produce materials—moving from extractive, energy-intensive processes toward growth-driven fabrication, where form emerges from the interaction between engineered biology and designed environmental conditions.

Giorgia Cannici is an Assistant Professor at Virginia Tech School of Architecture, working at the intersection of bioengineering and advanced manufacturing. Her research focuses on engineering living materials—programming organisms to produce and organize matter for architectural applications.
Trained across biology, engineering, and architecture, she holds a degree in Biology from Harvard University, a degree in Biomedical Engineering from Tufts University, and a PhD in Biological Systems Engineering from Virginia Tech. She is also a registered architect in Europe. Prior to academia, she worked at leading international architecture firms including Zaha Hadid Architects, Foster + Partners, and UNStudio.
Her work explores how biological processes can be harnessed as fabrication systems, enabling new approaches to sustainable, high-performance materials for the built environment.

The race toward AI-generated CAD has largely focused on teaching models to produce geometry directly—from B-reps and meshes to feature histories and sketches. But what if we’re teaching AI the wrong language?

This presentation explores an alternative paradigm: Text-to-Code-to-CAD, where large language models generate engineering logic rather than geometry itself. Instead of predicting low-level geometric entities, AI writes executable CAD code that captures design intent, constraints, parameters, and engineering relationships. The result is not merely a finished model, but a native, editable, and reusable engineering feature that integrates naturally into the design workflow.

Using real-world examples, including AI-generated CAD features built through FeatureScript, this session demonstrates how modern language models can write, test, debug, and refine engineering code to create complex parametric designs while preserving the transparency and editability required for professional engineering. The discussion contrasts this approach with direct geometry generation and examines why engineering intent—not geometry—may be the more scalable representation for AI-assisted design.

Attendees will leave with a new perspective on computational design: the future may not be AI replacing CAD, but AI leveraging programming languages purpose-built for engineering to accelerate design while keeping engineers firmly in control.

Darren Henry serves as Senior Vice President of General Operations at PTC, where he orchestrates diverse teams spanning marketing, technical services, customer success, education, documentation, and training. A degreed mechanical engineer, Darren has held leadership roles at SolidWorks, Atlassian, OpsGenie, InVue, and Copia Automation. An expert in modern product development practices, he brings over 30 years of experience helping manufacturers adopt new technologies to accelerate innovation, improve operational efficiency, and build better products.

“How can we develop better products, faster?” This is a fundamental question faced by every manufacturing industry. However, the traditional process, in which humans repeatedly carry out design, simulation, and testing, requires enormous time and cost and is increasingly reaching its limits amid intense global competition.

The rapid advancement of AI over the past few years has opened a new path beyond these constraints. From Design Optimization to Generative Design, we are now entering the era of Autonomous Design, in which Agentic AI can autonomously derive optimal design solutions. The use of AI in the design process is no longer optional, but essential, fundamentally transforming the paradigm of product development.

In this presentation, I will introduce the principles of AI-driven design, AslanX, Narnia Labs’ AI platform that brings these principles into practice, and a range of successful real-world applications across the manufacturing industry.

Namwoo Kang is the CEO of Narnia Labs and an Associate Professor at KAIST. He previously worked as a Research Engineer at Hyundai Motor Company.

He received his Ph.D. in Design Science from the University of Michigan. He also earned an M.S. in Technology Management and a B.S. in Mechanical and Aerospace Engineering from Seoul National University.

His research focuses on Agentic AI-driven engineering design through the integration of physics and data. His research interests include generative design, data-driven design, machine learning, deep learning, design optimization, topology optimization, CAD, CAM, CAE, and HCI.

Many Additive footwear platforms promise a level of customization beyond traditional footwear manufacturing processes, but how custom is it really? This presentation will explore a variety of strategies for the integration of various custom data into bespoke additive-ready algorithms in footwear design. The ability to create force-responsive structures through various types of micro lattice to tune flexibility, gradient behavior, anisotropy all within the confines of additive manufacturability parameters. Approaches include multi-agent algorithms with intelligent emergent behavior, simulation-driven anisotropic stochastic lattice structures, and body-responsive data mapping for truly one-of-one additive footwear creations.

David Burpee is a Computational Designer based in the Pacific Northwest, working across the footwear, apparel, consumer goods, automotive, medical, and architecture industries. He has lectured on Computational Design and Algorithmic Thinking at the University of Washington as part of a National Science Foundation grant exploring Engineered Living Materials. Over more than a decade, David has delivered Advanced Computational Design, innovation strategy, tools, and training for brands like Nike, PUMA, FILA, Brooks Running, General Motors, Harry’s Razors, and EQLZ. Originally trained as an architect with a Master of Architecture from USC, he has contributed to highrise and supertall projects in Los Angeles, Seattle, and across Asia. His work applies biomimicry, generative systems, and sustainable innovation to complex design and engineering challenges.

This talk presents a CAD-ready topology optimization framework for thermal management applications, including cooling channels, heat exchangers, and so on. By combining explicit geometric representations with physics-based optimization, the proposed approach directly generates manufacturable designs while preserving design intent and additive manufacturing constraints.

 Dr. Kentaro Yaji is an Associate Professor at The University of Osaka. His research centers on computational design and optimization of thermal-fluid and structural systems, integrating topology optimization and machine learning. He develops fast and adaptable design methodologies by fusing physics-based multifidelity simulations with data-driven modeling. His work spans theoretical advancements and industry collaborations, aiming to enable innovative and efficient product development.

This work presents an optimization workflow integrating Ansys SimAI (an AI surrogate model platform) and Ansys GeomAI (a shape generation AI model platform) for efficient concept exploration. 
Starting from a limited dataset, optimization is driven by SimAI surrogate models; however, when prediction accuracy deteriorates, GeomAI invokes solver evaluations to generate synthetic data for further SimAI Traiining. 
This training loop continuously augments the dataset and retrains SimAI, ensuring robust performance across an expanding design space.
The approach significantly improves optimization stability, data efficiency, and design exploration capability.

Ryoma Okamoto, Applications Engineering, Sr Staff Engineer at Ansys Japan K.K.
With a background in structural simulation and CAE software development, he focuses on AI/ML-driven acceleration of engineering analysis, automation of simulation workflows, and the integration of data-driven approaches into advanced simulation environments.

Aeroforms emerges from a shared curiosity of how the invisible sensory phenomena of scent can be modelled, mapped and ultimately understood as data, with a unique focus on scentscapes of Japan. 

Scent is one of the most powerful triggers of memory and belonging, yet it remains largely absent from environmental modeling and urban analytics. In 2001, Japan’s Ministry of Environment created “100 Scents of Japan”, a curated archive of culturally significant smellscapes tied to regional identity. The catalog revealed how smell functions as a vessel of ecological collective memory. Two decades later however, the physical landscapes that produced these scents have changed. Industrialization, land-use transformation, and air pollution have altered Japan’s atmospheric composition, and with it, its olfactory ecology.

Aeroforms surfaces at the intersection of environmental science and design engineering to address this gap by asking: how can we computationally model scent at a national scale, and what happens when we compare modeled atmospheric composition to cultural memory? 

We developed a computational framework for olfactory modeling, a way to approximate the composition of scents across Japan using quantitative environmental data. Through this framework, we constructed a sensory digital twin of Japan’s olfactory environment, integrating agricultural and industrial datasets to estimate the relative presence of ten scent taxonomies across all prefectures. We derived emission factors from literature-based odor threshold values using inverse-potency summation, median log compression, and adaptive normalization. These were then multiplied by prefectural land area to estimate natural emissions, with industrial pollutant emissions incorporated as a parallel weighted contribution layer. The result is a combined, normalized composition vector per prefecture. 

By comparing these modeled scent compositions with the “100 Scents of Japan” catalog, we measure where cultural memory aligns with — and diverges from — ecological and industrial reality. This research therefore operates at the intersection of environmental data science, cultural geography, and sensory humanities, developing a reproducible model for environmental perception and introducing a methodology for quantifying perceptual-environmental divergence or, the measurable difference between how people remember environments and how those environments now exist materially.

Zinab is a Master’s student in Design Engineering (MDE) through the joint Harvard Graduate School of Design and Harvard John A. Paulson School of Engineering and Applied Sciences program. She received her bachelor’s degrees in computer science and molecular biology from Brown University, focusing on human-centered interaction and neuroimmunology. Her research interests are all unified by a desire to make invisible systems legible, tangible, and experientially rich.

Esha is a Master’s student in Design Engineering (MDE) at Harvard Graduate School of Design and Harvard John A. Paulson School of Engineering and Applied Sciences. Her work aims to bridge complex, real world constraints with hardware and sensing solutions through design. She consistently pursues hands-on human centered approaches from developing interactive toolkits for women in Nigeria with the World Bank to building bidirectional sensing systems between search and rescue dogs and handlers in earthquake scenarios.

Lattice structures have become one of the most representative applications of additive manufacturing (AM) because of their lightweight characteristics, superior specific mechanical properties, and multifunctional capabilities. Although significant advances have been made in lattice design, process development, and topology optimization, these topics have largely been studied independently. A comprehensive engineering framework that systematically connects design, manufacturing, mechanical evaluation, and multifunctional optimization remains limited.
This presentation introduces an integrated framework developed through a series of studies on additively manufactured lattice structures. The framework first establishes optimal AM process conditions for producing high-quality lattice geometries. Theoretical methods are then presented for efficiently predicting the stiffness and strength of ideal lattice structures. To address practical manufacturing issues, experimentally quantified geometric imperfections are incorporated into structural evaluation, enabling realistic prediction of mechanical performance and providing robust design methodologies based on as-built geometries rather than idealized models.
The framework is further extended to multifunctional applications, including thermal-fluid management and acoustic engineering, where lattice architectures are optimized to satisfy multiple performance requirements. By integrating manufacturing quality, mechanics, imperfection-aware design, and multifunctional optimization, the proposed methodology bridges the gap between as-designed and as-built lattice structures.
The presentation demonstrates how this integrated approach provides a practical foundation for the reliable design and qualification of next-generation lattice structures for industrial applications while outlining future opportunities in digital qualification and physics-informed engineering.

Prof. Kuniharu Ushijima received his B.Eng. degree in Mechanical Engineering from Tokyo University of Science in 1997, his M.Eng. degree in 1999, and his Ph.D. in Mechanical Engineering in 2002 from the same university.
He joined the Faculty of Engineering at Tokyo University of Science as a Research Associate in 2002. In 2005, he moved to Kyushu Sangyo University as a Lecturer in the Department of Mechanical Engineering and was promoted to Associate Professor in 2008. During 2008, he was a Visiting Researcher at the University of Liverpool, UK. In 2014, he returned to Tokyo University of Science as an Associate Professor, and since 2019 he has been serving as Professor in the Department of Mechanical Engineering.
His research interests include additive manufacturing, lattice structures, structural mechanics, topology optimization, and multifunctional design. His recent work focuses on integrated engineering methodologies for additively manufactured lattice structures, covering process optimization, imperfection-aware mechanical evaluation, and multifunctional engineering applications.
He has authored more than 40 journal papers and has been actively engaged in research on the mechanics, design, and qualification of additively manufactured lattice structures.
He is currently preparing a book entitled Mechanics and Applications of Lattice Structures, which presents an integrated engineering framework for additively manufactured lattice structures.

The distance between a designer’s first sketch and a manufacturable product has traditionally been measured in weeks of handoffs: sketch, render, CAD, CMF spec, prototype. Vizcom is collapsing that distance.

In this talk, Jordan Taylor, co-founder and CEO of Vizcom, shares how AI-powered design tools are reshaping the pencil-to-product process for physical design teams in industry . He’ll cover how sketch to render generation evolved into a full creative pipeline, including region-based CMF editing tied to PLM systems, agentic design copilots, and the evaluation infrastructure needed to route creative work to the right models. The through line: rendering is no longer the end of the design process, it’s the beginning of production. Real is the new render.

Jordan Taylor is a transportation designer turned founder. He spent his early career in automotive and product design studios, including roles at Honda and Nvidia, where he saw firsthand how much of a designer’s time went to translation work rather than design: redrawing the same concept for the render, for the CAD model, for the CMF spec, for the review. In 2021 he co-founded Vizcom with Kaelan Richards to close that gap, starting with a simple idea that a sketch should be enough to get you to something real.

OPED (Optimization of Electric Drives) is an industry-approved software solution for the computational design of electric vehicle powertrains.

Starting from vehicle system requirements, the software automatically generates optimal powertrain systems, which consist of power electronics, electric machine and gearbox – each with numerous subcomponents to be designed. AI methods, namely evolutionary algorithms and artificial neural networks, are used to optimize the highly complex powertrain system regarding multiple design objectives – in particular cost, energy efficiency, performance, package, and sustainability. Best possible trade-offs between conflicting objectives are identified and the result is a Pareto front of optimal electric powertrain designs for the specific application requirements. Design engineers and decision makers are provided with this Pareto front within 24 hours and select the favored solution to guide the subsequent development.

The software is improving product characteristics and strongly reducing time-to-market, which is industry-approved in practical use at a leading global tier 1 automotive supplier and a world-renowned automotive original equipment manufacturer (OEM) – with high potential for scaling across others. OPED is demonstrated based on a real-world customer case study covering the design optimization of an electric powertrain for a passenger car application.

Martin Hofstetter studied Mechatronics and completed his PhD in the field of electric powertrain design in cooperation with MAGNA Powertrain. At the Institute of Automotive Engineering, Graz University of Technology, he developed an industry-approved AI-based design method for electric powertrains called OPED. The method optimizes for cost, performance, efficiency, package and sustainability, while also strongly reducing development time.

Currently, he is Head of E-Mobility and Alternative Drivetrains Research Group, focusing on application-oriented research in cooperation with industry partners. His primary goal is the commercialization of the developed OPED software in the course of a spin-off company . 

He is a frequent speaker at international automotive engineering conferences and his work is awarded by both industry and research institutions, including MAGNA, AVL and the German Association of Engineers (VDI). In the scientific community, he serves as reviewer for journals and conferences published by e.g., Springer Nature, Taylor & Francis, IEEE, FISITA and SAE.

To build the next generation of robots, we needed to build the next generation of design tools.

Software Defined Matter is a fully parametric, agent-first design substrate for high-performance machines. Geometry is implicit — signed distance fields — and purpose-built to feed optimization loops rather than simply representing CAD geometry. The resulting models are differentiable end-to-end, allowing gradient descent to run through requirements, geometry, physics, materials, and manufacturing constraints. Conventional CAD kernels were built to represent shapes. The SDM kernel is built for exploration.

A design is represented by an .sdm file: a graph carrying parameters, geometry, materials, interfaces, objectives, and the constraints the design must satisfy. From requirements through validation, all relevant information is traceable in one open, extensible object. Changes propagate rather than drifting between discrete tools. Manufacturing limits bound optimization from the start instead of being checked at the end.

Because design intent can now exist as one artifact rather than a disconnected pile of exports, the complete ecosystem can read and write the same source: optimizers, solvers, viewers, and manufacturing tools and specifications. SDM was initially built to design electromechanical systems for robotics, but we believe this approach is widely applicable to the design and manufacture of all physical things. The talk will cover how Software Defined Matter works under the hood, and how you can use it to change the way machines are designed and built.

Our computational design approaches to optimisation are often constrained by the tools at our disposal and by extension the established methods of generating new optimal forms.

With the revolution in AI coding tools, the barrier for designers to develop their own software tools and pipelines has never been lower.

Sub-Optimal explores what becomes possible when the context of optimisation is reframed around two priorities: (i) preserving the original designer’s macro intent, and (ii) embracing, through geometry, the anisotropic surface roughness that plagues additive manufacturing. The resulting approach is neither lattice application nor topology optimisation, but a simulation, surface texture and build orientation driven grown honeycomb structure that remains largely invisible at the macro level.

The talk closes on how the pipeline behind this approach was built in six weeks for $400 of AI tokens, leveraging a rich ecosystem of open-source libraries (including geometry3Sharp, PicoGK and OpenVDB) and formats such as 3MF. At a fraction of the cost of off-the-shelf tools, all that matters now is how creative you, the designer, can be.

Sarat Babu is a designer, engineer and strategist who has spent over fifteen years working at the convergence of computational design, materials and advanced manufacturing. He founded and scaled the additive technology consultancy Betatype, building London’s only commercial titanium additive manufacturing site before its acquisition in 2020; served as Chief Digital Officer at Alloyed; and most recently developed AI and AR wearable devices as a Principal Product Design Engineer at Meta. He holds a doctorate and several patents in medical, consumer electronics and additive manufacturing fields. He currently helps technically ambitious companies build products to match their vision, alongside an independent research practice exploring design and materials for product engineering, where he still builds hands-on.

Presentation Abstract

By now, implicit modeling technology and its benefits are well understood. In situations where a part is highly complex or its design process is highly iterative, implicit modeling using signed distance fields is uniquely able to deliver the reliability, performance, and scalability that are required for success. However, the novelty of implicit modeling is also a source of problems. Current engineering/manufacturing processes are often heavily dependent on legacy software tools that cannot consume implicit models — they require meshes or boundary representations, instead. So how do we address this problem? There are a few options — we can either convert implicit models to the forms needed by legacy tools, replace the legacy tools with new ones that consume implicit models directly, or change the data flow in the overall process. The talk examines these issues, reports progress and the current status, and suggests some ways forward.

George Allen has been a Technical Fellow at nTop (formerly nTopology) since 2019. Prior to that, he worked for the various owners of the NX/Unigraphics CAD system for four decades, holding various roles in R&D, strategic planning, and customer support. His main R&D contributions were in surface and solid modeling and user interface design. In addition, George consulted with major customers world-wide, helping them to use CAD technology to improve their product development processes. In the academic field, George has taught courses at SIGGRAPH conferences and the University of California. He is an associate editor of the CAD journal, and the author of a forthcoming book on geometry for computer applications.

The pressure to compress concept-to-product timelines reveals a critical flaw: the simulation stack cannot scale. Traditional pipelines are bottlenecked by geometry, which constantly changes representations across the lifecycle, moving from Implicit or CAD in design, to meshes, voxels, G-Code in manufacturing, to raw point clouds in QA. Because legacy analysis requires conformal meshing, manual preprocessing is forced at every stage, stalling automated iteration.

To evolve, simulation must become a geometry-agnostic infrastructure. Using immersed grid methods, Intact natively ingests these variable representations without preprocessing, exposing physics as an API-first service. We demonstrate how decoupling physics from geometry eliminates the manual bottlenecks, showcasing a workflow that handles diverse lifecycle formats to accelerate delivery timeframes.

This presentation outlines the development of an automated design workflow in an aerospace R&D project focused on heat exchangers manufactured using metal 3D printing (L-PBF). The project required extensive geometric exploration of the diaphragms in fluid channels by generating large numbers of 3D models by changing multiple parameters that define the geometry. Conventional 3D-CADs proved inefficient for this scale of design iteration, motivating the creation of a more programmable and scalable approach.
To address this challenge, a hybrid workflow was established using Grasshopper and nTop, two computational design platforms with fundamentally different geometric representation methods. Grasshopper, based on B-Rep method, enables detailed manipulation of non-thickness geometries but suffers from inherent mathematical limitations such as self-intersection during thickening operations. In contrast, nTop employs SDF (signed distance field/implicit modeling), which inherently avoid self-intersection and provide robust performance in thickening and boolean operations, though they are less suited for detailed geometric manipulation.
The presentation explains the contrasting characteristics of these two software systems and how to use them to compensate each other for their respective limitations. Through practical examples, it illustrates how the established workflow automates geometry generation, supports large scale parametric exploration, and enhances the efficiency of design iterations for complex AM components.

Founder of YAMAJI DESIGN / computational designer (nTop and Grasshopper) – He is an engineer with more than 15 years of experience in DfAM (design for additive manufacturing) across multiple industries. At a service bureau specializing in AM, he provided R&D support and DfAM training for U.S. military bases in Japan. He later worked in an aerospace company, where he contributed to R&D project to develop the heat exchanger for airplane using metal L-PBF.

Linkage mechanism design still requires extensive manual work across CAD and CAE tools and depends heavily on individual engineers’ expertise. Although LLM-powered CAD agents can accelerate autonomous CAD modeling, many current agents remain focused on part-level geometry and feature creation, with limited support for assembly-level mechanism design and engineering validation.
METHEUS addresses this gap by combining a proprietary generative mechanism model, engineering logic, and design ontologies. It guides engineers and automates execution throughout the design process, translating design intent into executable engineering requirements and generating, evaluating, and validating optimized mechanism designs through integrated CAD and CAE workflows.
Together, these capabilities transform CAD automation into an integrated engineering workflow, enabling autonomous mechanism design across mobility, robotics, and other linkage-based systems. By combining design intelligence with automated execution, METHEUS advances the AI transformation of mechanical design.

Jungho Kim is the Founder and CEO of IDeA Ocean Inc., where he develops AI-native engineering systems for mechanical design. He received his Ph.D. in Mechanical and Aerospace Engineering from Seoul National University, with research focused on autonomous mechanism synthesis methods.
His current work focuses on automating mechanical design by combining generative models, engineering logic, and accumulated design knowledge. Through increasingly autonomous workflows, he aims to shorten the path from an initial idea to a validated design and ultimately to physical realization.

Additive manufacturing is enabling a new generation of vapour chamber designs with geometries and functionalities not achievable through conventional fabrication methods. By leveraging AM design freedom, internal wick structures, fluid pathways, and enclosure geometries can be optimised to enhance thermal performance, reduce mass, and improve integration.
This presentation introduces our approach to additively manufactured vapour chambers, including a brief company overview and the relative patented associated.
Experimental results are presented to assess performance and repeatability, highlighting both the opportunities and current challenges of AM-based vapour chamber technologies.

Danilo Gigantelli is computational designer working in advanced thermal management and additive manufacturing, with a particular focus on vapour chamber technologies. His work centres on the development and optimisation of additively manufactured thermal devices, with an emphasis on understanding the relationships between process conditions, material behaviour, and device performance. Gianluigi Rossi and Humar Hossain will also likely be involved.

Generative AI increasingly allows designers to move directly from text to 2D images and 3D models. While this accelerates production, interaction often remains organized around prompting, generating, and selecting finished alternatives. This project asks a different question: what if AI generated not the design itself, but an evolving repertoire through which the designer could think—in other words, a design space that evolves through interaction?

AI Repertoires presents an experimental workflow for architectural ideation built around a persistent, AI-generated inventory of spatial and formal possibilities. Starting from a designer’s initial intentions, the system generates a repertoire of forms that can be rapidly selected, combined, retained, rejected, mutated, or regenerated. Rather than replacing the inventory with each generation, individual forms persist and evolve through interaction, allowing a project-specific visual and spatial language to develop on the fly.

The workflow is implemented through an agentic system in which reusable design skills coordinate form generation, combination, and interpretation, while tool connections enable geometry manipulation and architectural precedent retrieval. As forms are combined, emerging assemblies retrieve built precedents exhibiting related spatial or formal conditions. These references do not prescribe the next operation, but provide architectural feedback through which the designer can reconsider, retain, or redirect the evolving repertoire.

The project revisits the idea of architecture as an interactive repertoire, explored in early computational experiments such as Flatwriter and URBAN5, under a new generative condition: the repertoire itself can now evolve continuously through designer intention, emerging form, and architectural reference.

Jimmy Wei-Chun Cheng is Special Faculty in the Computational Design program at Carnegie Mellon University’s School of Architecture. His research investigates artificial intelligence, architectural representation, and human–machine design processes, with a particular focus on multimodal and agentic AI systems for architectural design. Prior to academia, he practiced architecture as a project architect with Junya Ishigami + Associates in Tokyo.

The traditional “design-to-analysis” loop is often bottlenecked by the laborious process of mesh generation, which can consume up to 80% of total simulation time. This presentation introduces Coreform IGA for Abaqus, a solution that brings CAD-exact isogeometric analysis (IGA) directly into the Abaqus ecosystem. By utilizing Coreform’s Isogeometric Analysis approach, users can perform high-fidelity simulations directly on CAD geometry, bypassing the need for traditional finite element meshing.

Matthew Sederberg, CEO of Coreform, has spent over two decades innovating at the intersection of geometry and physics. His work focuses on eliminating the “mesh bottleneck” through the commercialization of isogeometric analysis (IGA). Following the acquisition of his first company, T-Splines, by Autodesk, Matthew has focused on building Coreform to commercialize CAD-based simulation. He is passionate about creating workflows where geometry remains exact from the first sketch to the final simulation, a mission he continues to lead through Coreform’s integration with established ecosystems like Abaqus.

Computational design has outpaced simulation. TPMS structures, lattices and field-driven geometries can now be generated in minutes, but validating their thermal or structural performance still means days of geometry repair, meshing and solver setup, because conventional meshing tools were built for boundary-representation CAD, not implicit geometry. The result is a verification bottleneck: the most advanced designs are the least likely to be simulated before they are printed.

This talk presents a different approach: our meshing engine that works natively on implicit geometry, tool-independently, from any design source, for general simulation types, producing analysis-ready meshes for conjugate heat transfer and structural workflows in minutes rather than days, without geometry cleaning. We will show live workflows on additively manufactured heat-exchanger geometries, from implicit design straight through to converged CFD results, using industry-standard solvers.

We will then demonstrate how robust, deterministic meshing becomes the foundation for the next step: an AI simulation agent that orchestrates the full loop: meshing, case setup, convergence monitoring and reporting, compressing simulation setup from days to under minutes, with every engineering-critical parameter confirmed by the engineer rather than guessed by the model. Validated on production heat-exchanger cases, the approach shows how computational designers can iterate against real physics at the pace they design- reclaiming engineering hours without removing engineering judgment.

Drawing on deployments with aerospace manufacturers on both sides of the Atlantic, we close with lessons on what it takes to make implicit-geometry simulation routine in production environments.

Dr Liang Yang is the CEO and founder of Voxshell, a Cranfield University spin-out building ChopMesh — a patented meshing engine that makes implicit and additively manufactured geometries simulation-ready in minutes — and the AI Simulation Engineer, an agent that automates the full simulation workflow on top of it.

Before founding Voxshell, Liang spent over fifteen years in computational engineering research: a PhD at Swansea University, postdoctoral research at Imperial College London, and seven years as Lecturer at Cranfield University. His research spans mesh generation, computational fluid dynamics and fluid-structure interaction, with publications across leading journals in the field.

Tidal rotor blades present a challenging computational-design problem: parametric external geometry, internal composite architecture, hydrodynamic loading, structural limit states, fatigue life, and manufacturability must be resolved together. Yet preliminary structural design often inherits wind-turbine methods and their embedded box-spar assumptions, limiting exploration of alternative tidal-blade architectures.

This work presents two reduced-order engineering submodels developed for the EPSRC CoTide project—Co-design to Deliver Scalable Tidal Stream Energy: BladeStructurePrecompute and BladeStructureFatigue. Together, they couple an implicit parametric geometry kernel with structural screening and composite-fatigue assessment to support blade design before detailed analysis.

BladeStructurePrecompute is a reduced-order conceptual and preliminary-design package that parameterises the kernel to automate generation and evaluation of internal composite-blade architectures. Finite-element material-stress recovery enables concepts beyond the conventional box-spar assumptions embedded in legacy wind-turbine tools. The package currently represents 40 concept-level internal architectures. For a user-selected architecture, it generates laminate and geometric refinements and evaluates them against structural limit states; selection between architectures is intended to reside within a higher-level CoTide co-design controller coupled to other physics models.

BladeStructureFatigue is a reduced-order, ply-resolved composite-laminate fatigue limit-state assessment. Following technical discussions with the Bureau Veritas composites group, ongoing work implements the group’s composite-laminate damage-accumulation methodology within the CoTide fatigue package. Rainflow counting and cumulative-damage analysis resolve fibre- and matrix-dominated fatigue damage at ply level. Development is informed by laminate-fatigue research and full-scale blade testing undertaken at the state-of-the-art University of Edinburgh FastBlade laboratory, enabling accelerated lifetime testing to assess structural performance.

The doctoral work sits within the University of Edinburgh’s MATTERS group, whose architected-matter approach combines material and geometry across scales to tailor structural performance and resource efficiency. FastBlade connects this work to composite blade design, manufacture, static testing, and regenerative accelerated full-scale fatigue testing.

Sam Hughes is a postgraduate researcher in the Institute of Infrastructure & Environment, School of Engineering, at the University of Edinburgh; a member of the EPSRC Wind & Marine Energy Systems & Structures CDT; and part of the MATTERS research group based in the FastBlade laboratory. Sam’s PhD research develops reduced-order, parametric computational tools for the structural engineering of composite tidal turbine blades within the CoTide project. The work integrates implicit geometry, finite-element stress recovery, composite-laminate fatigue modelling, and multidisciplinary co-design, connecting digital design methods with composite manufacture and full-scale blade testing. Sam holds an MEng in Mechanical Engineering from the University of Edinburgh.

Additive manufacturing requires design workflows that connect geometry, simulation, and manufacturing data. This is especially important for lattice-based structures, where performance depends on complex geometry, effective material behavior, and fabrication constraints.

This presentation introduces Nodi, an integrated computational design platform for AM. Built around a CAD kernel and CAE solver developed for AM workflows, Nodi enables implicit modeling, lattice generation, mechanical analysis, and manufacturing-aware data generation within a unified environment.

Founder and developer of Nodi.
With a background in real-time computer graphics and 3D software development, he works at the intersection of geometry processing, computational design, and integrated CAD/CAE workflows for advanced manufacturing.

HYBEX Inc. is a startup developing and commercializing running footwear based on proprietary architected metamaterial structures.
Launched in 2026, our first product, HX LENS, combines a 3D-printed rotationally compressible lattice with conventional foam materials. While delivering performance comparable to modern high-performance running shoes, the system introduces a distinct running sensation by assisting the runner’s weight transfer throughout the gait cycle.
This presentation will explore the development process behind integrating a novel metamaterial architecture into running footwear. We will discuss the design rationale, computational design workflow, prototyping iterations, mechanical evaluation, and the challenges of translating an experimental architected structure into a manufacturable consumer product.