Introduction:

In footwear design, innovation rarely comes from following the beaten path. Pete Davis, Vivobarefoot's Computational Systems Lead, has crafted an incredible journey—from professional breakdancing to pioneering breakthroughs in computational design for footwear. In this interview, Pete shares how his unconventional background has shaped his fresh perspective on footwear, leading to groundbreaking projects at Vivobarefoot's innovation hub, BIOME.

With the introduction of the VivoBiome line, Vivobarefoot is pushing the limits of what custom-fit shoes can be. By leveraging advanced 3D scanning, AI, and parametric modeling, they are rewriting how shoes are designed, manufactured, and fitted. Pete dives deep into the computational tools and systems driving this revolution, from Grasshopper scripting to ShapeDiver's powerful scalability, as well as Volumental's advanced 3D scanning technology and Balena's innovative compostable materials, giving a behind-the-scenes look at the new frontier of personalized footwear.

This conversation with Pete Davis is a must-read for anyone interested in computational design, 3D printing, or the power of rethinking traditional industries. His insights highlight not only the challenges but also the immense opportunities in applying advanced digital tools to solve very human problems—like making a pair of shoes that fits just right.

1. What’s your role in the company? Can you tell us about your background and how it led you to innovate footwear design?

My name is Pete Davis, and I’m Computational Systems Lead at Vivobarefoot.

My career path has been quite unconventional. I left school to become a professional breakdancer, representing England for 17 years. After that, I transitioned into model making for architects and eventually founded my second company, specializing in computational design for 3D printing, AR, and VR. I sold the company to a client, ECCO Shoes, and helped establish their in-house innovation lab, ILE.

Not having any prior knowledge of footwear design actually worked in my favour, allowing me to approach the challenges with fresh eyes. Footwear is a fascinating field due to the deceptive complexity of designing around such malleable, force-dynamic, double-curved objects in multiple environments.

Unlike explaining the depth of computational design projects in architecture or aerospace, footwear is something everyone can relate to because we've all experienced discomfort from ill-fitting shoes. This makes the problem space instantly accessible and a fun problem to work on.  

Fell in love, married, and moved to Germany. I then focused on automating bespoke products through AI-based engineering, working for some leading companies that were working on new geometry kernels. This, with my past, caught the attention of VIVO, who was a client at the time. They brought me on board to lead their Computational Systems at BIOME (innovation department), where we’re pushing the boundaries of what footwear can be by leveraging parametric modeling and questioning traditional design norms. I'm excited to share more about our latest project today, which exemplifies this approach.

2. Can you tell us a bit about Vivobarefoot and the philosophy behind your minimalist footwear? How does Vivobarefoot's approach to footwear differ from other brands?

Vivobarefoot is a pioneering brand in the minimalist footwear industry. It is known for its commitment to creating shoes that promote natural foot movement and improve overall foot health. The philosophy behind Vivobarefoot's footwear is rooted in the belief that the human foot is inherently strong and capable and that traditional footwear, with its thick soles and restrictive designs, often hinders rather than helps the natural function of the feet. This thinking ensures its shoes enhance mobility, strength, and natural movement by incorporating several key design principles:

1. Thin, flexible soles for enhanced ground feel and natural flexibility.

2. A wide toe box that allows natural toe splay and improves mobility.

3. Zero-drop design that promotes natural foot positioning and reduces joint impact.

4. Lightweight construction that reduces fatigue and increases agility.

5. The absence of arch support or excessive padding to encourage foot strengthening and natural foot function.

6. Use of durable and responsive materials that are sustainable and adaptable to various terrains.

7. Educational resources to guide users in transitioning safely to minimalist footwear.

8. Continuous innovation based on biomechanical research and testing to refine shoe designs in alignment with natural foot function and the latest scientific findings.

3. How has introducing the VivoBiome line revolutionized the concept of custom-fit footwear? What did you improve compared to previous approaches?

The introduction of the VivoBiome line has revolutionized custom-fit footwear by utilizing three key pillars that are not normal in current footwear production.

1. Advanced 3D scanning technology,

2. Developments in manufacturing

3. A systems-driven computational approach to each foot.

If we start with just the traditional industrial production of shoes, every shoe is built around a Last. This shoe-shaped object is usually design or product-centric, leading to some aesthetically pleasing forms but functionally detrimental mechanics for the health of your feet.

This is evident in most sports shoes, with narrow toe boxes leading to toe compression problems later in life. Bunions are especially prevalent in women's feet due to their extremely high heel-toe shapes.

Vivo's main range of Lasts aims to address this with larger toe boxes and many other fit-form functions of the shape. At VivoBiome, we take this a little further, leveraging the data we can capture from thousands of foot scans and how we analyze and parse all the data.

For example, each person after scanning now has a VIVO foot health score. Within this score is an array of metadata we can filter from our data centers. We can pull up all scans within a certain heel-to-toe length and filter out if these people have low arches, bent toes, and minimal angle roll. We are left with a subset of scans that we use to build a new data-driven Last shape.

All of the Lasts made at VivoBiome are foot-centric shapes first. This is simple but exceptionally different for the footwear industry.

We aim to eradicate the use of Last altogether, which is unheard of. The computational system we have built allows us to do this, and I will elaborate on it later.

4. What role do advanced scanning and AI technology play in creating the VivoBiome shoes?

We have designed all workflows to be foot-centric, so the core of all algorithms needs to be the most accurate and user-friendly process. After evaluating the market for potential partners, we found that Volumental excels in many areas. They specialize in developing technology for 3D scanning and custom fitting solutions in footwear. Their seamless integration of fast, precise 3D foot scanning with AI-powered recommendations and insightful data sets them apart.

We have two ways to scan your feet: an in-store experience with bespoke hardware for us or software utilizing the latest in photogrammetry.

In both cases, the result is a mesh representation of each foot derived from thousands of scans using machine learning. We then orient and analyze the scan with our proprietary software.

5. How has using Grasshopper and ShapeDiver contributed to the innovation and customization of Vivobarefoot products?

As mentioned above, all analytics and foot health scores are parsed and processed using Grasshopper. This can all be done on a local machine if we are making custom lasts for a subset of people, but if we want to address making footwear for every foot, we need to think of a system that is so far removed from Last and human bottlenecks. Essentially, we need to automate the whole creative approach and democratize the creation process for our designers and, in turn, our clients. A great quote from Oren Harari is:

"The electric light did not come from the continuous improvement of candles."

The amendment of the Last was a new candle. The electric light bulb moment for VivoBiome was the creation of our patented process, which removes all human elements and automates fit for a digitally foot-centric approach. It all starts with how humans break down the workflow in traditional manufacturing.

Design intent comes first, and then a CAD engineer usually needs to realize this intent. The CAD engineer needs to talk next to manufacturing. The machinist will usually want amendments due to reality's limitations, and last, the material scientist's idea of what we can pump into the machines. Each stage creates human feedback loops in how they want to optimize or amend the true intent of the product. We start by distilling all the knowledge and opinions into code and flipping this hierarchy backward. Material, Hardware, Software, and Design.

We start with the constraints of how the material and hardware work in synergy, then feed these into robust software that the designer can always play in. The result is always a manufacturable product exported in native code for the chosen hardware. If we focus more on one of the workflows for the compostable Biome Moccs.

- Material: Balena BioCir 

- Hardware: FDM

- Software: GH + enhanced dependencies

- Design: Browser with ShapeDiver

The intent is an environment where the designer at vivo or client can analyze their foot and make design amendments without breaking the system. Export native gcode, never boundary representations. If you look closer, you will notice that for Balena, we use no Last but use data that the user can control to infer fit. Localized deformation that doesn't break the wall thickness of the constraints listed in material and hardware parameters. This equals almost shrink-wrapped footwear (not the Rhino 8 function) to your deformed fitted digital scan.

All of our shoes at Biome are created from just three images. These images affect elements like tread algorithms. This is majorly important to us, as accessibility allows all users to create, even our clients.

Again, extract the human element out of the process. The computational designer or CAD engineer can sometimes be the issue. They are so deep in understanding 3D that we can't bring every client up to speed on 3D design changes, but everyone can do 2D and draw on templates to make their perfect shoe. 

As mentioned in my experience above, I've worked on many geo kernels in the past, and sometimes, it feels like your hands are tied to the functions and environment in which you can work.

Grasshopper's development and maturity of functions are unmatched compared to other software. This allows us to think more deeply about what geometry we want to manipulate and transition between established geo types to achieve our goal, always keeping in mind the speed of execution of the script, as this can hinder a user's workflow.

Nodes are a great start, but they can slow all workflows down. We tend to program headless and incorporate multiple libraries to create an unbreakable workflow. Then, port this back to GH to leverage ShapeDiver for the front-end accessibility. With this new system thinking, the code we have written over the last few years has greatly affected Vivo. We can make design amendments in browsers via images or get clients to analyze their fit preferences and update the show around their feet in seconds.

6. What benefits did ShapeDiver's technology provide in developing the VivoBiome line?

We benchmarked many SaaS services for porting the workflow online. Three elements were really needed to ensure we could rely on letting our customers interact with the code we had created.

Security 

If you dare imagine the transition and protection of user data between multiple APIs and data centers, all our data is encrypted until the physical product is delivered. However, many SaaS offerings break this by keeping data on their end. With our ShapeDiver dedicated system, we are fully in control of how we utilize this cloud infrastructure.

Isolated environments for each user, minimizing the risk of data breaches and ensuring that proprietary designs and parameters remain confidential.

Computational Speed

We found that ShapeDiver significantly enhances computational speed by utilizing cloud-based processing to handle the complex calculations involved in running our scripts. Resulting in faster performance and the ability to manage more intricate and resource-intensive designs. The cloud instance can scale computational resources dynamically, ensuring that models run efficiently even when dealing with large data sets or complex vivo designs. A great example of this we used for our SaaS benchmark was a visual representation of marching cube iso meshes. These can usually result in slow render times, but I have no idea how SD has ported this so well to OpenGL.

Scalability

This is a big one. There was a limit to using our own servers or changing Rhinos Hops, and I didn't want to sink time into coding SD's offering. Queuing management spreads the processing of multiple scripts efficiently, ensuring that computational resources are used effectively while maintaining reliable performance. This is key if we have multiple users analyzing scans all at the same time. Load balancing and support of asynchronous processing are major bonuses for us. In short, ShapeDiver sped up our product to market.

7. Apart from VivoBiome, was ShapeDiver useful for tackling other challenges at Vivobarefoot? 

The products we have released are really the tip of the iceberg. Many I can't talk about yet, but all are ported to SD and accessible via an online script for in-house and Vivo customers.

One story that SD really impacted is the development of the Last production.

Normally, footwear is more of an art than a science, so all last measurements are done using physical methods that can result in mixed results due to the non-planar nature or rotation of the Last shape. Biome’s task was a simple script: take two lasts and compare the differences between them, measurements, etc. This could be a great little script on a local machine, but really, I, along with the ego of CAD/computational engineers, was the issue.

I'm the experienced bottleneck for vivo colleagues and the Last manufacturer to make sure we are getting what we want.The script was ported to a test website and used SD Iframe for all the parameters. It was exported as a PDF with cut-throughs and multiple key data points. The result was that non-3D people scrutinized our Lasts and accelerated the process of getting to the best Last for normal business at Vivo.

8. Can you describe the customer experience from scanning their feet to receiving their custom-fit VivoBiome shoes?

After signing up at our store or from the comfort of your own home, you will be guided seamlessly on one website through each stage we need to get your custom Biomes.

First, scan. If you are scanning from home, all you need is an A4 sheet of paper. This data is then stored on our back-end system and linked with our next step.

We did all web development UI in-house but linked it to ShapeDiver's back-end power. We call this stage the fit visualizer. Here, you can see your scans and assess the health measurements of your feet. Then, you can flick between different pre-made designs or traditional footwear that we sell.

With each choice, you can see how your foot fits in the shoe in full 3D. We also use Grasshopper to deform the shoe to fit the foot scan based on the material of each part of the shoe, something I've not seen before.

From here, users can choose what design or parameters they want to be changed on the bespoke footwear, such as tread types, logo design, or localized fit amendments.If the user has chosen Balena or our conformal offerings that are fully 3D printed, then the user can amend all elements on their foot. If you want to make a sandal, amend the input image template, and bang, you have it. Clients' cost calculations and environmental impact scores all update accordingly. Finish, and all files will be sent in native code to our data center and then to localized production.

9. How do you ensure that the 3D-printed shoes maintain both performance and appearance? Can you discuss the materials and technology used in creating your compostable shoes?

I will jump back to the Material, Hardware, Software, and Design philosophy.  

Once all deformation changes have been made to the scan, we build the shoe from there using multiple techniques.

Material offsets concerning Balenas bio material BioCir are always multiples of the FDM nozzle diameter used, and some changes that I won't go into.

As we are working at such a low level in regards to geometry linked to open vdb fields, we can manipulate the temperature of the nozzle in Gaussian local transitions across one slice of the build. Higher temperatures create a more rigid area; lower temperatures, for example, are used on the upper of the shoe.

The result is a uniformly offset wall thickness from the foot with meta changes that affect the performance of the footwear. Balena’s BioCir is a groundbreaking bio-based material designed to degrade naturally in soil, providing an eco-friendly alternative to traditional plastics. Made from renewable, non-toxic ingredients, BioCir breaks down quickly under natural conditions, converting into carbon dioxide, water, and biomass without leaving harmful residues.

This allows Vivo to create new business models to leverage the product's life cycles. We also have the option to send the shoes back, and we can remake the new pair or design one from the same material.

At vivo we started to think and implement, “Made to be Remade”.

10. What impact do you hope the compostable shoe will have on the footwear industry and environmental sustainability?

The compostable shoe made with Balena’s BioCir material has the potential to revolutionize the footwear industry by setting new standards for sustainability and innovation. By replacing traditional, long-lasting plastics with BioCir’s compostable, bio-based material, the industry can significantly reduce plastic waste and lower its carbon footprint. Footwear being one of the largest contributors to waste in the design industry.

This shift supports the transition to a circular economy, where products are designed to return to the environment safely and efficiently. Additionally, BioCir's use encourages broader adoption of eco-friendly materials, appeals to environmentally conscious consumers, and helps preserve ecosystems by preventing soil and water pollution. These long-chain polymers are really prevalent across materials for 3d printing, so to see companies like Balena spearhead this movement is great. Overall, this advancement could drive meaningful change in how footwear is designed and manufactured but also inform the customer that this is possible and affect change for customer-driven demand. Vivos clientele are usually environmentally minded, so this has already affected the intent of what we want to achieve and amend with this product. 

11. How do you see the future of footwear evolving with the integration of 3D printing and custom-fit technology?

We are one of many people trying to crack this problem. People have been really failing fast since the mid-2000s for 3D printed products for footwear. This knowledge has really shaped our approach, but many of the core pillars of our patented process have been from failing in very granular areas of the workflows that unlock light bulb moments that make everything else easier.

A good example of this is the orientation of the foot scan before all other functions. Seems simple, but usually, the simplest things for humans to do are the hardest in code. We now have a robust methodology that was inspired mainly by shoemaker patents from the 17th century. A lot of companies are offering 3D printed elements, inlays to a full shoe. The majority of these are pre-designed objects in a range of foot sizes, again all human-driven problems that the design has to jump through to make it printable. Mostly made in CAD/VR iteration until it's printable. This is kinda like the traditional method of making footwear anyway, but just printed. So when mentioning a company does fit, it's just fit on a small premade range level. it’s never linked or built just from a foot scan. If a foot scan its just matching your dimensions to work that's been done before. In my opinion, custom fit should be tailor-made for you, with every stitch calculated and every tread component linked to your data. Biome, in my opinion, is this now! Such an exciting time.

The best part is our method is only strengthened by the software paradigm. When we amend each update, all fits will only get more accurate.

We are agnostic to the input scans or the output for the native code, so quickly changing how the new material will act to fit with design changes is achieved at a rapid pace in contrast to normal production. In essence, every element in our hierarchy is linked to the best barefoot fit you can achieve with your data. 

12. What’s the future of the Grasshopper ecosystem in the 3D printing space?

Grasshopper has so many good plugins for 3D printing but a Design for Additive Manufacturing mindset is more what's needed. This can really shape how you create your products and software that relates to it.

Really, if you are using most geometry software, you are bending what it was initially designed for. You are manipulating boundary representations, but real-world objects aren't made of boundaries; they're volumetric. 3D printing allows us to make volumetric objects, but the kernels we all tend to use don’t even support this.

Grasshopper has some great implicit and SDF libraries, but they are never linked to the machine's native code creation. So there is always the step of exporting to a mesh for slicing, or some subtractive machinists will insist on a .stp or .iges file as this is what they are used to, and they want to amend the file. This breaks most workflows for custom products and can never be automated. What we are doing at Biome is always linked to other constraints of how the machine makes the product, which in turn makes the design space a certain way. Plus, sending native code to the printer means no human interaction with the customer's file. We have floated the idea of making a Vivo plugin for Grasshopper to address some of these problems. GH is widely used in footwear, but I would love to see McNeel keep pace with some of the new SDF / Implicit kernels that are on the market or the work that Fusion360 has done on design-to-print.

Shrinkwrap is a great function in the latest update of Rhino, and it already uses OpenVDBs to make it work. So, is this a hint that Voxel could be in Rhino 8? If so, linking native code would only be a stone's throw away.

- This has been fantastic. Thank you very much for your time!

Thank you very much as well, and thank you to the ShapeDiver team for helping us throughout this process.

That's it for this new edition of Getting to know... Don't forget to visit VivoBiome's page and follow Vivobarefoot on all social media channels!

Would you like to get featured in this space? Please email us at contact@shapediver.com and tell us about your project or brand! We'd love to start a conversation.