CDFAM Computational Design Symposium

Duann Scott
  • 5,0 (1)
  • DESIGN

CDFAM Computational Design Symposium Presentation Recordings www.designforam.com

  1. -8 H

    Intelligent Anatomic Models from CT Utilizing ML - Matt Shomper - Not a Robot

    Organization: Not a Robot Presenter: Matt Shomper Recorded at CDFAM Computational Design Symposium, NYC, October 29-30, 2025https://cdfam.com/nyc-2025/ Intelligent Anatomic Models from CT Utilizing ML Presentation Abstract This presentation discusses an accessible system that takes CT scans and automatically turns them into detailed 3D models while intelligently tagging important anatomical features. Instead of engineers and researchers spending hours manually creating these models and identifying landmarks, our approach uses machine learning to do the heavy lifting. The process works by feeding CT scan data through specialized algorithms that can recognize the structures and convert the flat scan slices into three-dimensional representations. At the same time, the system automatically identifies and labels key anatomical points like bone structures or tissue edges – creating a smart, annotated 3D map of what was scanned. This has the ability to dramatically speed up workflows that previously required tedious manual work. The automated tagging means that medical professionals get consistent, standardized labels across different cases, which is especially valuable for surgical planning and patient-specific implants. The presentation will cover some challenges of utilizing M/L, how manual inputs can train algorithms over time, and looking towards the future of validating such systems for true use in commercialized systems. Speaker Bio Matthew is a visionary leader in the computational design of advanced 3D-printed medical implants, with close to 15 years of experience in engineering, research, and innovation. As an inventor, creator, and passionate leader, he has been a part of founding businesses focused on additive manufacturing and is an internationally recognized speaker on biomimicry, computational modeling, and additive manufacturing – lecturing at conferences and prestigious universities including MIT and Harvard. Matthew’s work is driven by his passion for exploring the macro and micro of biological forms, turning algorithms into functional structures for physical devices. He has pioneered the idea of a “biologically advantageous implant,” and has also spearheaded multiple public initiatives to synthesize biological structures as computational models for use in engineered products. He currently is the founder and principal consultant of Not a Robot Engineering, a co-founder of LatticeRobot, and involved in several other stealth startups. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    17 min

  2. -1 J

    AI and the Battle for the Soul of Design - Chris McComb - Carnegie Mellon University

    Organization: Carnegie Mellon University Presenter: Chris McComb Recorded at CDFAM Computational Design Symposium, NYC, October 29-30, 2025https://cdfam.com/nyc-2025/ AI and the Battle for the Soul of Design Presentation Abstract Artificial intelligence is reshaping the landscape of design and additive manufacturing, accelerating creative workflows while challenging long-held assumptions about authorship, originality, and human intuition. As AI becomes more deeply embedded in computational design tools, it offers unprecedented capabilities for exploration, optimization, and customization—often revealing solutions that elude traditional design methods. Yet this power comes with profound questions: What does it mean to design when machines generate ideas? How do we preserve the human element in a process increasingly influenced by algorithmic reasoning? This presentation examines emerging patterns in AI-driven design, the shifting role of the designer, and the ethical dilemmas that arise when intelligence—natural and artificial—co-create. Through examples from additive manufacturing and beyond, it offers a vision for navigating this new design frontier without losing sight of the creative soul at its core. Speaker Bio Chris McComb is the Gerard G. Elia Associate Professor of Mechanical Engineering at Carnegie Mellon University. His lab, the Design Research Collective, advances interdisciplinary design research by merging perspectives from engineering, manufacturing, psychology, and computer science. He also serves as the Director of the Human+AI Design Initiative, an interdisciplinary and international group of researchers focused on application of human-AI collaboration to design, with support by industry partners. He is affiliated with the NextManufacturing Center, the Manufacturing Future Institute, and the Wilton E. Scott Institute for Energy Innovation. His research interests include human social systems in design and engineering; machine learning for engineering design; human-AI collaboration and teaming; computation for advanced manufacturing; and STEM education. He received dual B.S. degrees in civil and mechanical engineering from California State University-Fresno. He later attended Carnegie Mellon University as a National Science Foundation Graduate Research Fellow, where he obtained his M.S. and Ph.D. in mechanical engineering. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    20 min

  3. -1 J

    Design as Dialogue: Form Jamming with AI Agents - Matthew Goldsberry & Junling Zhuang - HDR

    Organization HDR Inc Presenters: Matthew Goldsberry & Junling Zhuang Recorded at CDFAM Computational Design Symposium, NYC, October 29-30, 2025https://cdfam.com/nyc-2025/ Presentation Abstract While AI is often used for visualization in architecture, its potential to directly generate and shape geometry within the design process is still emerging. This presentation explores how we have been integrating model-aware AI agents into our design process. We begin with Synthesizer, a custom browser-based modeling tool paired with an Arduino-powered physical controller. Through a simple physical controller, designers trigger higher-order parametric actions, making the act of modeling feel more performative than procedural. Our early beta experiments, focused on building minimal, controller-driven interfaces, explore new possibilities beyond the traditional mouse and keyboard. We then introduce Form Jamming, a method developed within our RhinoMCP workflow. It treats the initial burst of AI-generated geometry as provisional material—something to be shaped and refined into architecture through intentional, iterative moves. While still experimental, this approach has shown promising results in several recent projects, a few of which we will share. This work outlines a new model of computational authorship in which designers and AI agents collaborate through structured dialogue. It points toward a future where generative design is not only more contextual and adaptive but also legible, editable, and deeply integrated into the design process through natural language interaction. Speaker Bio Matthew GoldsberryMatt oversees the applied research and implementation of advanced computational design workflows. He is the director of Data-Driven Design and is responsible for developing new computational tools and workflows to facilitate design exploration, automated analysis, and advanced data management. Matt is also a Lecturer at the University of Nebraska-Lincoln, where he teaches courses on advanced geometry and building information modeling. Matt holds a Master of Architecture degree from the University of California Los Angeles and a Bachelor of Science in Architecture degree from the University of Nebraska-Lincoln. Junling ZhuangJunling is a design technologist bridging research and practice in the AEC industry. As a software engineer at HDR’s Data-Driven Design team, he develops AI-powered 3D tools. Junling holds an M.S. in Computational Design from Columbia and is pursuing an M.S. in Computer Science at Georgia Tech. His work has appeared in ACADIA and CAADRIA, and he reviews for top venues including ACADIA, CAADRIA, TAD, and FoA This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    22 min

  4. -1 J

    Superintelligence for Scientific Discovery in the Material World - Markus J. Buehler - MIT

    We open Season 5 of the CDFAM podcast with a keynote presentation from the CDFAM Computational Design Symposium NYC 2025 by Markus J. Buehler, McAfee Professor of Engineering at MIT. In this session, Buehler outlines how AI is becoming an autonomous partner in scientific discovery—capable not only of analysis but of generating new knowledge. Drawing on examples from materials science and bioengineering, he presents multi-agent AI systems designed to reason, hypothesize, and evolve—creating a framework for discovery engines that extend the boundaries of human-led research. Buehler’s work brings together reinforcement learning, graph-based reasoning, and physics-informed generative models, with case studies showing applications in medicine, food, agriculture, and beyond. Organization: MIT Presenter: Markus J. Buehler McAfee Professor of EngineeringMassachusetts Institute of Technology Presentation Abstract AI is rapidly transitioning from a passive analytical assistant to an active, self-improving partner in scientific discovery. In the material world, this shift means developing systems that not only recognize patterns but also reason, hypothesize, and autonomously explore new ideas for design, discovery and manufacturing. This talk presents emerging approaches toward ‘superintelligent’ discovery engines -integrating reinforcement learning, graph-based reasoning, and physics-informed neural architectures with generative models capable of cross-domain synthesis. We explore multi-agent systems inspired by collective intelligence in nature, enabling continuous self-evolution as they solve problems. Case studies from materials science, engineering and biology illustrate how these systems can uncover hidden structure-property relationships, design novel materials, and accelerate innovations in medicine, food, and agriculture. These advances chart a path toward AI that actively expands the boundaries of human knowledge in engineering. Speaker Bio Markus J. Buehler is the McAfee Professor of Engineering at MIT and a pioneer in AI‑driven knowledge discovery. He created powerful graph‑aware, multi‑agent AI platforms that turn heterogeneous data into science-grounded actionable insight, powering breakthroughs in materials science, biology and healthcare. Buehler is among the world’s most‑cited materials scientists and the recipient of numerous honors, including the Feynman Prize, ASME Drucker Medal, J. R. Rice Medal, and the Washington Award. He is a member of the U.S. National Academy of Engineering. For more than a decade he has also taught executive and technical professionals at MIT, shaping the next generation of leaders in engineering, knowledge discovery, and artificial intelligence. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    30 min

  5. 16 OCT.

    Bioinspired And Biobased 4D-Printing For Adaptive Building Facades - Tiffany Cheng - Keynote Presentation

    Organization: Cornell University Presenter: Tiffany Cheng Bioinspired And Biobased 4D-Printing For Adaptive Building Facades Presentation Abstract What if our buildings and products could be manufactured and operated the way biological systems grow and adapt? As an alternative to conventional construction and manufacturing, I will present a bioinspired approach to making through material programming and 4D-printing. By integrating material, structure, and function, plants change shape over varying spatial-temporal scales in response to external stimuli. I will introduce how computational fabrication enable the bioinspired interplay of cellulosic materials, mesostructures, and adaptive motions to create hygromorphic systems powered by the environment. The developed methods are transferable across scales and applications – from hobbyist 3D-printers to industrial robot platforms and self-adjusting wearables for the body to weather-responsive shading in buildings. Through integrative technologies and interdisciplinary solutions, we can leverage biobased materials and bioinspired design principles to create a built environment that is transformative and resilient. Interview: Bioinspired and Biobased 4D-Printing for Adaptive Building Facades – Tiffany Cheng Tiffany Cheng is a Taiwanese American designer and builder whose work examines the performance potential of natural and biobased materials for smarter and more sustainable forms of making. As Assistant Professor at Cornell University’s Department of Design Tech, Tiffany directs the MULTIMESO Lab to develop computational fabrication processes for creating bioinspired systems across scales, from self-forming furniture to adaptive building components. Previously, Tiffany was Research Group Leader at the Institute for Computational Design and Construction (ICD) at the University of Stuttgart, where she led the Material Programming research group and earned her Doctorate in Engineering. Tiffany holds a Master in Design Studies (Technology) from Harvard University and a Bachelor of Architecture from the University of Southern California. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    32 min

  6. 13 OCT.

    Building Surrogate Models for Physics Simulation using a No-Code Approach

    Organization: Key Ward Presenter: Asparuh Stoyanov Building Surrogate Models for Physics Simulation using a No-Code Approach Presentation Abstract This project demonstrates a no-code methodology for building surrogate models for engineering simulation. Using such methods, physics simulation analysts can tap seamlessly into the potential of surrogate models, transforming traditional simulation workflows to be more efficient and flexible. In this abstract, we present a workflow of how to use simulation result data to build a 3D surrogate model that any analyst can utilize without requiring programming skills—enhancing the usability of AI-driven simulation tools for broader adoption. Finite Element Method (FEM) simulations are often computationally intensive and challenging to scale, especially for complex structural applications. Our methodology minimizes these resource-heavy processes with a graph-based surrogate model optimized for computational efficiency. To achieve this, we utilized automated extract, transform, and load (ETL) workflows to process the raw simulation data into a shape and format suitable for AI ingestion. We show how, through no-code data processing automation, analysts can focus on deriving insights rather than getting lost in technical details. The dataset used comprised linear static analysis results of a Press Bench model, performed using SOLIDWORKS Simulation. Parametric variables included back height, feet width, and plate length, and the results predicted were displacement and stress. Using data processing and management tools, we first extracted and converted the surface field and volumetric field data, from the original raw format into an open-source “AI-ready” format (. csv,.vtk). This allowed us to gather all simulation data in one place to better understand the data distributions, patterns, and correlations between variables. In the next step, we cleaned the collected data while maintaining different data versions and keeping track of changes. As a final step, using the cleaned and processed dataset, we trained a Graph Neural Network. The model was trained to predict accurate stress and displacement fields within seconds (>90% accuracy), using the 3D volume mesh data as inputs. The whole process from raw data to a trained model took approximately one workday to develop. The same approach will be tested on large deformation nonlinear structural analysis. This project demonstrates how structural simulation data can be used to build surrogate models that accelerate the design process. Advances in AI modeling tools now make these models widely accessible, enabling engineers to leverage physics simulation data without coding or deep machine learning expertise—expanding the possibilities in product design optimization. RECENT INTERVIEWS & ARTICLES * AI Judges in Design: Kristen Edwards – MIT * Manufacturing Driven Design with Rhushik Matroja – CDS * Beyond Surfaces: Applying Intrinsic Geometry Processing in Art and Design: Math Whittaker, New Balance * Maia Zheliazkova – On LightSpray * Design for Additive Manufacturing at CDFAM – Part 2: 2024 Berlin * Design for Additive Manufacturing at CDFAM – Part 1: 2023 NYC This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    18 min

  7. 2 OCT.

    Stress-Based Design Of Lightweight Horizontal Structures For Concrete 3D Printing - Luca Breseghello

    Recorded at CDFAM Computational Design Symposium, Amsterdam , 2025 https://cdfam.com/amsterdam-2025/ Organization: DTU Presenter: Luca Breseghello Stress-Based Design Of Lightweight Horizontal Structures For 3D Concrete Printing Presentation Abstract Concrete is one of the most widely used materials in construction, but it’s also a major contributor to CO₂ emissions. In mid-rise buildings, slabs and beams alone account for over 40% of the concrete used. This raises an important question: how can we build these elements more efficiently while reducing their environmental impact? In this talk, I’ll share how robotic 3D Concrete Printing (3DCP) and structural optimisation can work together to create lighter, more material-efficient beams and slabs. By integrating computational design, Finite Element Analysis (FEA), and stress-based material placement, we developed a workflow that reduces waste while maintaining strength. I’ll introduce 3DLightBeam and 3DLightBeam+, beams with double the strength-to-weight ratio of conventional 3DCP beams, and 3DLightSlab, a ribbed slab designed for efficiency. Structural testing and Life-Cycle Analysis (LCA) confirmed that this approach can lead to more sustainable concrete structures. This presentation will explore the practical potential of 3DCP in structural applications and what it means for the future of concrete construction. Interview: Stress-Based Design Of Lightweight Horizontal Structures For 3D Concrete Printing – Luca Breseghello – DTU Join us at CDFAM, October 29–30, to connect with the people defining the future of computational design. Not just the speakers on stage, but the researchers developing new algorithms, engineers scaling workflows into production, architects rethinking building systems, and designers pushing the boundaries of products and materials. CDFAM is where leaders and practitioners from across industries come together, sharing methods, exchanging ideas, and building collaborations that carry far beyond the event itself. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    20 min

  8. 29 SEPT.

    How Topology Optimization And AM Can Create A New Generation Of Green Steel Construction

    Recorded at CDFAM Computational Design Symposium, Amsterdam , 2025 https://cdfam.com/amsterdam-2025/ Organization: University of Bologna Presenter: Vittoria Laghi How Topology Optimization And Additive Manufacturing Can Create A New Generation Of Green Steel Construction Presentation Abstract The digitalization of the construction sector could potentially produce more efficient structures, reduce material waste and increase work safety. Current strategies for the realization of automated steel constructions see the application of metal 3D printing processes as an opportunity to build a new generation of efficient steel structures with reduced material use. This, though, requires advanced multidisciplinary knowledge in manufacturing, metallurgy, structural engineering and computational design. Recent effort has been made in order to combine computational design with current digital fabrication procedures to realize efficient steel structures for the future. The present work aims at providing insights to current explorations on the combined application of computational design and metal 3D printing process in construction towards a new generation of optimized and resource-efficient structures Interview: How topology optimization and additive manufacturing can create a new generation of green steel construction This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com

    19 min
Tout afficher (115)

À propos

CDFAM Computational Design Symposium Presentation Recordings www.designforam.com

Informations

  • Création
    Duann Scott
  • Années d’activité
    2023 - 2025
  • Épisodes
    115
  • Classification
    Tous publics
  • Copyright
    © Bits to Atoms
  • Site web de l’émission
    CDFAM Computational Design Symposium