CDFAM Computational Design Symposium

CDFAM Computational Design Symposium — Washington DC 2026

Brandon DeMille · General Atomics Aeronautical Systems

Computational design methodologies, including topology optimization, are transforming airframe structures development by enabling rapid exploration of design configurations during early conceptual phases. This presentation demonstrates a workflow that enables informed decision-making across disciplines and accelerates the path from initial concept to detailed design. A fuselage case study illustrates the simultaneous optimization of composite laminates for skins, substructure geometry, and overall shaping. This integrated approach facilitates quantitative trade-offs among competing priorities such as cost, structural performance, manufacturability, and production rate.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search this talk's transcript

CDFAM Computational Design Symposium — Washington DC 2026

Chris Helmerich · Celedon Solutions

Most ‘AI for engineering’ tools today sit beside the design process with a chat window next to a CAD viewer, a copilot that summarizes documents someone still has to act on. The harder problem is putting AI inside the loop, where it can read and write the same structured representation of the system that engineers, simulations, and downstream manufacturing all depend on.

This talk covers how we approached that problem at Celedon Solutions while building Davinci, an engineering platform where agentic AI operates directly on a live knowledge graph of the system under design. Requirements, components, interfaces, behaviors, and their relationships all live in one connected structure, and the agents that work on it can pull structured model content out of reference documents, generate and compare architectural alternatives against performance and cost constraints, trace requirements through simulation results, and reach into external tools such as parts databases, Python simulations, PLM systems through APIs and Model Context Protocol.

I’ll walk through the design decisions behind a graph-native rather than document-native foundation, show where generative exploration has meaningfully compressed early-phase trade studies in aerospace and defense pilots, and talk honestly about the failure modes where agents confidently produce plausible-looking nonsense, and what guardrails and iteration strategies actually work. The goal is a practical view of what it takes to move AI from the margins of the engineering workflow into the part of the process where design decisions actually get made.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search this talk's transcript

CDFAM Computational Design Symposium — Washington DC 2026

Jan Vandenbrande · nTop

Current Computer Aided Design systems excel in static detailed design but are too fragile and slow to support Design Exploration and Multidisciplinary Design Optimization for conceptual and preliminary design. This talk introduces a new approach to modeling products that overcomes these shortcomings based on implicit functions popularized in the animation industry. The main benefits of the approach is that is responsive to the need to design and redesign products in days or weeks and not month or years because of absolute robustness to parametric change minimizing human intervation; lightning fast evaluations leveraging GPUs; and performing analysis directly from the representation w/o the need of human intervention to generate cumbersome and error prone meshes.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

William Steadman · Quanscient

We present the results of the largest CFD simulations to date deployed on IBM and IonQ quantum hardware through our ongoing collaborations across the aerospace, maritime, and automotive sectors. We will explore the critical trade-offs quantum computing introduces to computational design in aerodynamics and aeroacoustics, while demonstrating how these advancements are being integrated into Quanscient’s multiphysics software.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

Austin Herrema · Istari Digital

As the Department of Defense accelerates adoption of digital engineering and advanced manufacturing, the challenge is no longer defining the digital thread—it is executing it across a fragmented Defense Industrial Base (DIB). This session will explore a consortium-based approach to demonstrating an end-to-end digital thread spanning design, build, operations, and sustainment—executed within each partner’s native environment.

Rather than forcing tool or data standardization, this effort enables participating organizations to use their own systems, data architectures, and processes while securely sharing only what is necessary to maintain a federated, authoritative source of truth. The result is a practical model for interoperability that reflects real-world constraints: multiple vendors, distributed ownership, and varying levels of digital maturity.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

Rhushik Matroja · Cognitive Design Systems

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.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search this talk's transcript

CDFAM Computational Design Symposium — Washington DC 2026

Daniel Hambleton · Metafold

Generative AI is rapidly expanding what designers and engineers can create in 3D, but visual plausibility is not the same as geometric fidelity. This presentation asks a practical validation question: how close are AI-generated CAD models really to a target desired shape?

We introduce a feature-vector workflow using the Metafold Shape Similarity technology to compare generated models against reference targets. Each model is encoded into a geometric feature vector, enabling direct comparison through aggregate similarity scores, coordinate-level distance ribbons, scale-normalized metrics, and side-by-side 3D previews. The result is a repeatable method for moving beyond “looks right” evaluation toward measurable shape correspondence.

Using examples from current 3D generative design workflows, the talk demonstrates how feature vectors can expose where a generated model preserves intent, where it drifts, and which geometric features contribute most to the gap.

This approach offers a lightweight validation layer for AI-assisted CAD: fast enough for iteration, interpretable enough for engineering review, and concrete enough to support model benchmarking.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

TJ Root · InfinitForm

The gap between optimized geometry and manufacturable components has been the defining constraint of computational design for three decades. Topology optimization produces brilliant forms that machinists cannot cut. Those forms are not editable in CAD / or CAD friendly. Simulation validates performance that mainstream manufacturing cannot reproduce. The result: design cycles measured in months, not days, and engineering organizations forced to choose between what is optimal and what is buildable.

InfinitForm was built to eliminate that tradeoff. The platform takes geometrical, engineering, manufacturing and cost constraints as input and outputs production-ready parametric CAD geometry, optimized simultaneously for structural performance and the specific manufacturing process it will be produced with, whether CNC machining, additive manufacturing, casting, extrusion, or injection molding. Every output carries full design history, constrained sketches, and parametric relationships, making it immediately editable in the CAD environment the engineering team already uses. GPU-accelerated solvers and optimizer, the system compresses what previously required weeks of iteration into minutes of compute.

This talk presents the technical architecture behind that capability, the manufacturing constraint modeling approach that makes outputs buildable rather than merely optimal, and results from production deployments at aerospace, defense, and advanced manufacturing organizations. It examines what changes when design for performance and design for manufacturing are solved as a single problem rather than sequential steps, and what that means for the engineering organizations, defense programs, and industrial supply chains now entering the Physical AI era.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

Sergey Pigach · CORE studio | Thornton Tomasetti

CORE studio spent a decade building machine learning tools for structural design and analysis, all running as cloud services behind APIs. When MCP arrived, handing those same tools to an agent turned out to be close to trivial. Sergey Pigach demonstrates Bender, an agentic system running on AWS that the firm talks to through Slack: ask it for a concrete column stack and footing for a five-storey residential building in New York, let it make the remaining assumptions, and it calls the tools the engineers use, renders the result and writes a design summary. It lives in Slack deliberately, because an agent sitting in a shared thread already has the context of the conversation around it, which a one-to-one chatbot does not. Specialist sub-agents handle questions like embodied carbon.

From there the talk moves to agents talking to each other. A2A is a protocol for delegation between agents, complementary to MCP rather than competing with it, but it has no discovery layer — a public agent the team put online was found by nobody. That gap prompted Waggle, Pigach's own side project, which crawls for valid agent cards and builds a searchable index with health, quality and trust signals, then delegates a request to whichever agent can handle it. He also shows agents paying each other small amounts to cover expensive work.

The most uncomfortable result is a benchmark. CORE studio asked its own engineers for the hardest structural problems they could devise, assembled 91 of them, graded answers to within one percent, and gave the models nothing but a calculator and a Python sandbox — no internet, no engineering software. They expected around half. It saturated immediately, with the leading models above 94 percent. Tracing backwards showed the unlock was reasoning: the first reasoning model jumped from 38 percent to 74, and the line has run straight up since. Structural engineering, as he puts it, is a verifiable domain. The talk closes on CAD experiments, including a Grasshopper plugin that exposes a parametric definition to an agent as MCP tools and a hackathon robot arm driven by natural language, and on the conclusion that this is not a domain expertise problem but an unhobbling problem.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

CDFAM Computational Design Symposium — Washington DC 2026

Rik Baruah · Intact Solutions

AI is reshaping engineering as it drives demand for surrogate models, large design studies, and agentic workflows that require automated simulation loops at scale. These pipelines collide with two realities: geometric representation is fragmented throughout the hardware lifecycle (from CAD to point clouds), and traditional FEA, reliant on conformal meshing and manual preprocessing, treats human intervention as a core requirement. This brittle paradigm resists automation.

Scaling simulation for the AI era is fundamentally an infrastructure problem. Using immersed grid methods, Intact natively ingests any geometry representation without preprocessing, exposing physics as an API-first callable function. We demonstrate how this architecture powers the emerging “engineering-as-code” stack through automated DOE pipelines and LLM orchestration via MCP across concept, manufacturing, and deployment.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation