In this week's newsletter, Aaron Moncur has a conversation with Duann Scott, founder of cdfam, a global symposium connecting engineers and designers in computational design and additive manufacturing. Scott leads the 3MF Consortium and previously held leadership roles at Autodesk and nTopology.
The design process and the manufacturing process in a very abstract way, completely linked like that is inseparable. This is kind of like a very organic, abstract kind of almost art approach to this, which can then change the way we think about how we design an injection molded part, perhaps.
In this episode:
Why computational design differs from parametric CAD—bringing multiple data sources together to solve design and manufacturing constraints simultaneously
How major footwear companies hire architects for computational design teams because they learned these workflows creating building facades
Why multi-objective, multi-stakeholder problems create opportunities for cross-industry collaboration between medical devices, aerospace, and architecture
How the 3MF file format addresses limitations in STL files for additive manufacturing data transfer
Bonus Content:
Lego’s mold design strategy
S6E37 Duann Scott | Computational Design & The Best File Format for 3D Printing
Duann Scott explains how computational design extends beyond traditional parametric CAD by building systems that integrate performance requirements, manufacturing constraints, and simulation data into a single decision-making platform. Scott describes how this approach enables medical device engineers to automate bone graft design from scan data, how footwear companies apply architectural design workflows to create customized shoe structures, and why a robot feeding mycelium over two months demonstrates the inseparable relationship between design and manufacturing processes. The conversation covers practical applications across industries, the role of grasshopper as the backbone of computational design, and why engineers transferring 3D printing data should adopt the 3MF format instead of continuing with STL files.
>If YouTube isn’t your thing, check out this episode and all of our past episodes on Apple, Spotify, and all the rest.
296 engineers. 154 training sessions. 4.2/5 rating. Zero catastrophes.
Our first full-scale PDX Conference is officially in the books, and the numbers tell a great story. Nearly 90% of registered attendees actually showed up (trade show veterans know that's unusual). Over half arrived between 9-10am ready to learn, and 84% said they're coming back next year.
The booth-centered training format - where exhibitors delivered live technical sessions right from their booths - created the kind of energy we hoped for. Engineers stuck around. Conversations happened. People made genuine connections instead of just collecting vendor swag and leaving.
Coming Soon:
Training session replays
Video testimonials from attendees
Raffle drawing winners announced here first
Thanks to everyone who made this happen: the PDX staff, Pipeline team members, keynote speakers (Jake Kennington, Leon Zhang, Aaron Eden & Ron Higgs), and the engineers who helped spread the word. And of course, our training partners who delivered 154 sessions that kept people engaged all day.
One attendee message stuck with us:
"I hope you feel the pride that you deserve. It made a big difference... It makes me motivated and encouraged and inspired that one person can do big things if they decide to do it and put their mind to it."
We learned plenty that we're already implementing for PDX 2026. The community showed up, engaged genuinely, and gave us clear direction on what worked. That's exactly what a first-year conference needs.
Next year will be bigger, better, but provide the same impact as we were able to accomplish this. See you next year!
Why Footwear Companies Hire Architects: Computational Design Skills Transfer Across Industries
Scott describes how computational design skills developed in one industry directly apply to entirely different fields, creating unexpected career paths and cross-pollination of ideas.
The backbone of computational design is really has been, for the past 10-15 years, has been grasshopper within the rhino platform. It's a very open system which is affordable at a student level and has been taught in architecture for over 10 years. And so there's thousands and thousands of people who know, who have learned, particularly in architecture, how to programmatically design systems to make things.
This creates an unusual talent pipeline where architectural training becomes valuable in manufacturing contexts.
If you think of how - the major sportswear footwear companies that you all know, love and wear, hire architects as part of their computational design team to build these systems. Because they know these computational workflows from doing building facades 10 years ago, and from facades to footwear, you think it's not that big a jump.
Using Their Legal Team Exclusively: Accelerators that insist you use their attorneys for documentation create inherent conflicts of interest. Their lawyers represent the accelerator's interests, not yours, in structuring deals and protecting your rights.
Don't think, oh, I'm going to trust them and I'm just going to go and we'll do the documents with their attorneys and it's like, no, don't do that.
The transferability emerges from working with similar abstract problems at different scales.
What I sort of can find is how to get people to work together from broad fields, to solve when they have an abstracted version of similar problems. When someone from the medical device space being called in to work with someone in the aerospace industry - if someone did a presentation on how osteo integration may happen, and then they're using that same idea for filtering chemicals to help with a chemical reaction. That's cool. That's what's interesting, like when it's abstracted out.
Lego’s Mold Design Strategy
LEGO's mold design decisions directly enable their business model: tools must last 30+ years while producing billions of elements at consistent quality. The company maintains over 3,700 active element designs but rarely introduces new ones because each new element requires substantial capital investment, ongoing maintenance costs, and permanent inventory complexity that must be justified against decades of production economics.
For more, visit the full article on The Wave.
This week Dale wonders…
I was wondering what programs everyone might be referencing to start implementing formal ANSI Y14.5/ ISO 2468 (or the newer standard) in their shops without say a full iSO 9001 certification?
I've seen some of the trainers on the podcast, and I'm wondering:
1. What kind of metrology tools the recommend purchasing for self training?
2. If there are managably priced trainers or coaches who work with staff to get them involved? (Not to mention, how to get people motivated?)
3. Can anyone tell me about their successful implementation in a company that had long used general dimensioning schemes?
Thanks!
Want to chime in? Visit our forum on The Wave.