On the manufacturing side, practical minded engineers ask themselves why designers often don’t listen to them. The former head of engineering for Nike prepared a complete engineering guide for footwear that he compiled using everything he learned across his career – it was his legacy before retiring, filled with an incredible wealth of testing information and best practices. But it was a static and external resource, something seen as suspicious and outmoded by the design teams meant to adopt it, and it wasn’t used.

Software struggled from the same tension between design and engineering until the advent of no-code platforms allowed designers to build websites and workflows themselves. Finally designers were able to not just build for themselves, they learned the capabilities and limitations of the software they were building within and understood where they could get creative. Websites became more beautiful as the level of design elevated across digital properties.

And it wasn’t just designers who benefited. By focusing on building tools that designers would use, engineers began to understand design and its process better, tailoring capabilities accordingly. Their task went from executing on a design to building better tools for designers to create with.

No-code platforms broke down the translation barriers between two critical teams, design and engineering, and allowed each to better understand and work with the other.

Now we are focusing on going from no-code to no-CAD, allowing designers to build physical products and toolchains themselves. This builds on the transformation no-code platforms brought to software, extending it to physical product creation and onto the factory floor. No-CAD platforms enable designers to create manufacturable 3D products without the years of training and experience required with CAD software or specialized product engineering.

In the example above, a designer moves from developing an initial concept to redlines to building out the outsole and developing a custom texture for its sidewall within a few minutes. The outsole is generated alongside 2D cut-views of each section so that the designer can make changes in 2D and have them impact 3D space. What formerly would have been a hand off and led to a technical design package and then a 3D model is now driven entirely by design.

We’ve learned that the key to an effective no-CAD platform is the ease of translation between 2D and 3D space, physical sketches and models to digital ones - designers are creative across mediums and agile by nature, where one process is helpful to edit in 3D another is more intuitive in 2D.

Just like no-code platforms did for software designers and developers, no-CAD platforms raise the level of design across product creation while building better engineering and manufacturing teams. By centralizing and constantly updating a platform’s engineering standards with each prototype and launch, every test and trial, a developer’s experience is immediately amplified across design teams and built on by other teams in real-time. These platforms are more than just translators across teams – they become the living institutional memory and mind of an organization’s product creation engine.

One of the most powerful capabilities these platforms enable is the seamless training and integration of custom AI models within the design workflow. Whether it’s testing or training data, brand DNA, collaborations or archives, providing design teams with enterprise-level intelligence directly in their workflow will unleash a radically new generation of product.

Unlike no-code platforms, no-CAD platforms are only made possible through the latest leaps in computer vision and genAI. Over the past 18 months the team at HILOS has built customizable and modular AI-driven manufacturing workflows for footwear design that show the degree of creativity and control no-CAD platforms can deliver.

We began by focusing on dissolving the barriers between physical design mediums and digital product creation, allowing attendees at Vogue Business’ Fashion Futures to sculpt with clay and see 3D shoes change on screen through real-time computer vision capture.

At TechCrunch Disrupt we moved further into sketching, showing how a designer went from sketch to on-foot sample during a live 3-minute demo on-stage.

As we’ve built up our own no-CAD application, Interplay, and continued to develop our underlying operating system and 3D engine, we’ve forced ourselves to be comfortable with showing work when it’s still early and being open about our own conceptual growth as this fast-moving frontier expands. These ManufacturingAI systems are very new and promising, and they come at a time when how and where we make things is more important than ever.

For a more technical deep-dive into HILOS OS, read the white paper below.

Still curious? To learn what manufacturingAI means for the future of footwear, click here.

Or if you’re an executive at a large brand exploring how you can bring this into your company, we have a few tips for you here.

Not enough for you? Comment and subscribe below, or reach out directly.

Leave a comment

Subscribe now

Generative AI models are progressing from generating 2D images to 3D models, enabling them to begin learning not just how to represent design direction in 3D space but how to incorporate manufacturability into their results. The short-term impact of this progression is faster and more accurate product development as design for manufacturing (DfM) logic is automated into the design process. In the long-term, we believe this progression will see a shift from vertical to horizontal and composable manufacturing models - just like we see today with software development - that emphasize learning and integration over control and standardization.

Manufacturing applications each have their own specific constraints, criteria, and logic: how shoes are designed to be made varies greatly from apparel or eyewear. The logic for each application is learned and evolved over a career and varies significantly across individuals and companies. It’s also always changing with the introduction of new materials and manufacturing technologies. Teams perennially struggle to define and update this logic or to leverage it during the design process.

ManufacturingAI treats CAD as code, training AI models on engineering data in order to synthesize a set of manufacturing constraints that are continuously improved and evolved. 3D assets can be generated, updated, and improved in a networked, modular approach just as quickly as code. Product generations that normally take 2-3 years to develop, test, commercialize, learn, and update a given product line are set to shrink to weeks and months while delivering new design and performance capabilities to creative teams.

In the first weeks and month of the COVID pandemic we caught a first glimpse of this. CAD files for PPE were open sourced across the web, downloaded, printed, improved, and shared by a global community. One improvement in New Jersey was immediately felt in California, delivering product improvements at a blistering pace.

That same approach is now being seen in AI for robotics, where what one machine learns from a cycle of movements can be deployed across the fleet in real-time, accelerating group learning and rapidly improving robotic automation.

This approach is most extreme in Ukraine, where drone manufacturing mirrors this horizontal, interoperable approach and where product generations are being launched every three weeks.

Similar to the shift from vertical to horizontal that Microsoft brought to computing in the 1980s, digital manufacturing is entering a Cambrian moment that will deliver an explosion of new technologies and applications revolving around operating systems that are generative and composable while being hardware and application agnostic.

Tech Stack

HILOS is building an industry-specific OS for ManufacturingAI, adopting a horizontal approach that focuses on the specific logic for translating design inputs for footwear into actually manufacturable shoes that can integrate across a range of technologies.

Our OS is frontend and application agnostic, allowing for a wider diversity of hardware integrations and software applications, where improvements in one layer of the stack contribute to new capabilities and applications elsewhere.

1. Application layer

Standalone applications built on HILOS OS can be cloud-based or local, on-prem or off, and incorporate custom tools and third party plug-ins and integrations. Interplay is our own application and defined interface for those who may desire custom tools, models, and workflows but do not need a custom frontend, built specifically to incorporate the flexibility of custom tools and third party integrations.

Developer platform

Our SDKs allow third-party developers to build custom plugins at the application layer, rapidly expanding capabilities and features across the platform. This development community can create proprietary tools only available to a specific customer or openly available tools and plug-ins that are available for free or low-cost to the entire community of users.

2. Operating system

Our composable OS acts as a data unification layer that gives teams the freedom to create any footwear product with new or existing data in a single location. This source of truth provides teams the power to orchestrate and manage how product is designed and manufactured within the frontend experience.

3. Virtualization

Our serverless architecture allows developers to build and run applications without managing server infrastructure. In this model, cloud providers handle server provisioning, maintenance, and scaling. Applications are event-driven, executing code in response to specific triggers, and operate on a pay-as-you-go basis, where costs are incurred only when the code is running. This approach simplifies development and resources are automatically scaled to meet demand.

4. Model layer

Our Model layer combines general LLMs with specific language models trained on footwear design and manufacturing data, allowing for the rapid creation of custom models based on proprietary data streams. Proprietary data and hyperlocal models can be kept on-prem, ensuring ownership of all core IP and resulting AI capabilities.

Deep Dive into Operating System: MagiCAD

• MagiCAD

  Our underlying language for 3D asset creation, built over nerbs, mesh, and solid geometry kernel libraries. Language is an apt metaphor for how MagiCAD works: it classifies CAD assets as primitives, each made up of arguments (such as last, biteline, outsole curve) and properties, such as the color or material of that manufacturable asset. Primitives act as words that behave according to their own underlying grammar and syntax, allowing them to understand which other words they go with, how, and what that in turn creates.

  

  From a machine learning perspective, this linguistic foundation becomes even more powerful, as each parameterized “word” not only tells us how components fit together, but also translates those components into quantifiable data for analysis. By encoding 3D geometries as bits, numbers, and vectors, MagiCAD enables the extraction of feature sets ripe for training predictive models, automating workflows, and supporting intelligent design recommendations. This data-driven approach bridges the gap between geometry and computation, allowing for pattern recognition across entire product lines, optimization of material usage or performance metrics, and even the generation of novel designs through generative adversarial networks or other AI-driven techniques. In turn, this synergy between structured parametric data and machine learning accelerates innovation while preserving the intuitive, linguistic logic of the original 3D asset creation process.

• APIs

  Model agents then work to translate these words and sentences into discrete manufacturing technologies via APIs. This translation would be prohibitive without the use of agents that can take structured and unstructured data and learn from the translation process across different formats and manufacturing techniques.

Deep Dive into Application layer: Interplay

Interplay showcases our approach to generative, modular, and composable applications built for ManufacturingAI. The application is built as an infinite canvas waiting to be filled with different nodes, each one representing a different function or tool developed over HILOS OS. Designers can start with existing workflows or build their own, chaining various nodes together to deliver powerful, new product capabilities.

Nodes follow a general classification:

1. Hooks: These bring in outside data, either manually letting a user bring in an image or sketch, or automatically using webhooks or APIs to bring in real-time customer, testing, or cultural data. This includes custom AI models, or avatars, trained on both structured and unstructured data.

2. Generators: These generate assets, be they visual renders, cut views, or high-res 3D files for manufacturing.

3. Modifiers: These introduce specific capabilities to the user, modifying assets accordingly. Examples can be red-lining an existing render to change it, generating custom textures that load into a 3D texture pen for modifying a 3D asset, or changing lasts and morphing the assets accordingly.

With a modular approach, these nodes can be assembled into any order the user desires and are infinitely compostable. This has been uniquely well-suited to both the flexibility and customization designers demand while also delivering a developer platform to easily build and launch new nodes and capabilities.

Custom Model Building

This feature within Interplay is surprisingly powerful, allowing users and teams to quickly build custom models with a small amount of sample data or with webhooks and APIs continuously bringing in new data.

By introducing multiple model types—such as generative models for style exploration and surrogate models for performance optimization—Interplay enables a truly adaptive, data-driven design process. Image libraries, for instance, can be leveraged to train generative networks that produce new footwear concepts aligned with a brand’s established aesthetic or functional requirements. Meanwhile, test fit measurements and performance data can feed into surrogate models that predict how various design tweaks might affect comfort, durability, or other key metrics, streamlining the prototyping process. By combining these specialized approaches under one cohesive platform, Interplay empowers teams to continually refine their footwear designs with high fidelity, harnessing the user’s existing data to unlock faster innovation and usher in more personalized product experiences.

To learn what manufacturing AI means for the future of footwear, read on below:

Or if you’re an executive at a large brand exploring how you can bring this into your company, we have a few tips for you below:

Comment and subscribe below, or reach out directly.

Leave a comment

Subscribe now

ManufacturingAI is a profound change to both how we design and how we manufacture, allowing for designers to translate their design intent directly into manufacturable 3D while integrating real-time intelligence into the design process.

ManufacturingAI treats CAD as code, allowing for the same rapid improvements we are used to seeing in software to reach the factory floor. 3D assets can be generated, updated, and improved as quickly as code, while the translation of 3D assets into coding language leverages the rapid learning and inference capabilities of genAI. Key to this is the integration of massive amounts of data in real-time directly into the design and manufacturing chain, allowing for the same network advantages that software development enjoys, with new snippets of code, features, or functions developed across a community to quickly increase the potential of a given application.

The benefits of ManufacturingAI are a much higher diversity of specialized, rapidly improving product categories without the risk and waste of legacy development cycles. Product generations normally take 2-3 years to develop, test, commercialize, learn, and update a given product line; with ManufacturingAI these product generations are set to shrink to weeks and months instead of years while delivering new design and performance capabilities.

To my knowledge the only place on Earth where we have actually seen this in operation is in Ukraine, where drone manufacturing mirrors this horizontal, interoperable approach and where product generations are being launched every three weeks.

Similar to the shift from vertical to horizontal that Microsoft brought to computing in the 1980s, digital manufacturing is entering a Cambrian moment that will deliver an explosion of new technologies and applications revolving around operating systems that are generative, modular, and composable while being hardware and application agnostic.

The First Intelligent OS for Footwear Creation

HILOS is building the first industry-specific OS for ManufacturingAI, adopting a horizontal approach that focuses on the specific logic for translating design inputs for footwear into actually manufacturable 3D shoes that can integrate across a range of technologies. Our OS is frontend and application agnostic, allowing for a wider diversity of hardware integrations and software applications, where improvements in one layer of the stack contribute to new capabilities and applications elsewhere.

Application layer - Interplay

Standalone applications built on HILOS OS can be cloud-based or local, on-prem or off, and incorporate custom tools and third party integrations. Interplay is our own application and defined interface for those who may desire custom tools, models, and workflows but do not need a custom frontend.

Model layer

Our Model layer combines general LLMs with specific language models trained on footwear design and manufacturing data, allowing for the rapid creation of custom models based on proprietary data streams. Imagine designers building custom models around proprietary design archives or real-time testing data. Proprietary data and hyperlocal models can be kept on-prem, ensuring ownership of all core IP and resulting AI capabilities.

Developer platform

Our SDKs allow third-party developers to build custom plugins at the application layer, rapidly expanding the set of capabilities and features at the application layer. This development community can create proprietary tools only available to a specific customer or openly available tools and plug-ins that are available for free or low-cost to the entire community of users.

Operating System

• MagiCAD
  Our 3D engine supplies the underlying language for 3D asset creation, translating code into CAD that reflects the manufacturing logic learned by our models and the design intent of the user.1

• APIs

  Our APIs translates 3D CAD files into discrete manufacturing technologies: an upper is 3D knitted while paneling is 3D printed on an MJF, the bottom unit extruded using FDM. That a single design generated into a cohesive set of 3D assets can be discretized into a number of different manufacturing outputs is the critical sort-and-push approach used by our 3D processor and its surrounding APIs. This translation would be prohibitive without the use of genAI agents that can take structured and unstructured data and learn from the translation process across different formats and manufacturing techniques.

Our Manufacturing AI application: Interplay

Interplay showcases our approach to generative, modular, and composable applications built for ManufacturingAI. The application is built as an infinite canvas waiting to be filled with different nodes, each one representing a different function or tool developed over HILOS OS. Designers can start with existing workflows or build their own, chaining various nodes together to deliver powerful, new product capabilities.

Nodes follow a general classification:

1. Inputs: These bring in outside data, either manually letting a user bring in an image or sketch, or automatically using webhooks or APIs to bring in real-time customer, testing, or cultural data. This includes custom AI models, or avatars, trained on both structured and unstructured data.

2. Generators: These generate assets, be they visual renders, cut views, or high-res 3D files for manufacturing.

3. Modifiers: These introduce specific capabilities to the user, modifying assets accordingly. Examples can be red-lining an existing render to change it, generating custom textures that load into a 3D texture pen for modifying a 3D asset, or changing lasts and morphing the assets accordingly.

With a modular approach, these nodes can be assembled into any order the user desires and are infinitely compostable. This is uniquely well-suited to both the flexibility and customization designers demand while also delivering a developer platform to easily build and launch new nodes and capabilities.

To learn what manufacturing AI means for the future of footwear, read on below:

Or if you’re an executive at a large brand exploring how you can bring this into your company, we have a few tips for you below:

Comment and subscribe below, or reach out directly.

Leave a comment

Subscribe now

Hosted by ALCOVA1 at the Villa Bagatti Valsecchi2 for Milan Design Week 2024, we transformed a wing of this historic villa into an exhibition space that showcased a new collection of wearables produced under Studio, our innovation lighthouse where ideas are brought to life with brands that exemplify the future of craft.

The Exhibit

“It’s an Italian villa! We need more romance.” said Lillian Hardy, our art director. With Cooper Gill at the helm along with versostudio and the rest of our amazing team3, a freewheeling effort was made to design an experience that read like a love letter from earth.

Shoes emerged from dunes as a nod to volcanology and Fire of Love. A double arch colonnade bifurcated the room, with the back-of-house inspiring technical curiosity through various touchscreens and touchpoints. Attendees were encouraged to play with objects and materials and test drive Studio OS for the first time.

Our Vision: Born from Powder

A thin layer of powder4 is spread across the bed, selectively sintered with hot light. This process repeats itself layer by layer, thousands of layers at a time, tuned at the voxel level, until a recognizable form emerges from the dust. Like a buried artifact, the printed object is excavated from TPU powder and thoroughly cleaned in preparation for finishing and assembly. This method of 3D printing is called Powder Bed Fusion, and it’s a mesmerizing process. Christy Lai, who leads software at HILOS, calls this powder “shoes in the raw”.

To juxtapose: have you ever seen an active volcano throw pyroclastic ash into the air? This fallen ash, or tephra deposit, is the raw material of creation. It’s a journey from the center of the earth to the liberated surface as additive poetics of form. Our landscape has been printing 24/7 since the beginning of time, and with it, co-authoring every chapter of the human condition onto geologic record. We too, are printed from these particles, or as Carl Sagan would say: cooked in the hearts of stars. The volcano is earth’s original 3D printer.

The dunes embody the analog of modern creation, and our mission was to spark a dialogue using the material that connects us to Earth and to each other. It’s still awe-inspiring to imagine that a pile of sand or TPU powder could somehow transform into your favorite wine glass or 3D-printed sneaker. While nature is a true inventor, it takes human ingenuity to invent the tools that transform these primitives into wearable art, and human responsibility to ensure our methods are sustainable for a greener future. And this is where Studio OS comes in.

Hello, Studio OS

During our time in Milan, we unveiled a new product creation tool to the public. Studio OS is the first generative design software to leverage AI models trained on 3D footwear, turning concepts into wearable realities with uncompromising speed, flexibility, and control.

Learn more about Studio OS

Subscribe now

MARGINALIA

ABOUT JAMES

James Carnes is an Innovator, Brand Strategist, and Executive. In his 25+ years in the industry, he has launched concepts like Boost, Primeknit, and Futurecraft, and led adidas sustainability strategy for many years. James was born and raised near Detroit, Michigan.

LINKS

• HILOS website

• James Carnes on LinkedIn

With this project, Ancuta Sarca marks a creative expansion into the world of 3D footwear, putting a completely new stamp on how the world views luxury footwear design.

This is the first 3D-printed heel that combines harder nylons and softer TPU printed within one assembly, opening up a completely new range of design possibilities that still embody the performance and comfort that 3D promises.

Beyond the bold and modern styling of the footwear line, this launch showcased the speed of the new HILOS Studio platform, going from first conversation to launch in just four months.

"Although there are no new Swoosh silhouettes this season, there is room for a selection of 3D-printed minimalist sandals that arrive in vibrant red and blue made in partnership with HILOS Studio, recognized for bridging advanced technology with eco-conscious craftsmanship."

— Highsnobiety

“These are the first shoes that are made by blending two printed materials," said Sarca. "They work on a zero waste process, where they create everything digitally. So there is no waste."

As a part of London Fashion Week 2024, Ancuta Sarca and HILOS co-hosted an event at The Hoxton with Culted. See the afterparty here.