FLAVE exists to close that gap.

Built on top of more than 22 years of manufacturing experience, FLAVE is a company building on-demand production infrastructure for the sportswear industry. It’s not a brand. It’s the platform layer that lets anyone design, sell, and produce performance apparel without ever touching a factory floor.

Who it’s for

The core customer is the creator or emerging brand who wants to design and sell without touching manufacturing, plus established brands looking for on-demand capacity. Over 900 brands signed up ahead of launch, which tells us the appetite for this is real.

Designed to reduce waste

Traditional apparel is a forecast-and-warehouse business. You guess what will sell, produce it in bulk, and hope. What doesn’t move becomes deadstock.

FLAVE makes production genuinely on-demand, with global fulfillment and local shipping, so you produce what’s ordered, near the buyer. No speculative bulk runs, no seasonal overproduction, no deadstock, minimal freight. Waste isn’t managed after the fact. It’s designed out at the model level.

Empowering designers, not replacing them

There’s a fear that AI in design means cutting jobs. We see it differently.

The FLAVE design studio turns a plain-language brief into design options and a 3D preview, then exports production-ready files automatically. What it removes is the tedious file-prep and tech-pack drudgery, not the creative judgment. Designers iterate faster and skip the parts that were never the creative work in the first place: colour separations, spec sheets, file export.

It also lowers the barrier for people who couldn’t previously access manufacturing at all. That expands the market rather than shrinking the workforce.

Where it gets interesting: the garment becomes a platform

This is where we aim to truly transform the industry.

We’re building new technology directly into the garments, so that a print on a garment basically becomes a trackable barcode. During production, it gives live info on the production status to the customer. When the garment is ready, the same print is scannable with a phone.

From there, the print can trigger an AR advertisement or any other kind of interaction with the person who scanned the jersey, allowing a high ROI for sponsors. The garment stops being a static object and becomes a living touchpoint.

There is plenty more we’re exploring and testing right now. The best way to stay informed on the latest module releases that can be connected to FLAVE is to follow us.

Because FLAVE is more than just an AI design tool that makes print files. It’s the operating system for any sportswear brand that wants to play a role in the future.

The barriers we’re solving

AI adoption in this industry has stalled for real reasons: trust in output quality (will the AI file actually run on a real line?), fragmentation between design tools and actual factories, legacy workflows and tech-pack habits, and the gap between a pretty render and a production-ready file.

Most tools stop at the mockup. Bridging to genuine manufacturing is the hard part, and it’s exactly where we’ve focused.

Changing the business model itself

By decoupling selling from manufacturing, FLAVE lets a brand launch with zero inventory and zero production overhead, test designs live, and only make what sells. That turns apparel from a forecast-and-warehouse business into a demand-driven one. It lets small brands operate with reach previously reserved for large ones: design, store, marketing, order routing, fulfillment, and service in one interface.

Where this is heading

The future is an agentic operating system for brands, with AI handling design, routing orders to the nearest capable production, and managing fulfillment automatically across a distributed network. Print production becomes a responsive, on-demand node rather than a fixed offshore batch operation.

The full agentic platform is our target for end of 2026 or early 2027.

The garment is about to become intelligent. The supply chain is about to become invisible. And the brands that move first will define what the next decade of sportswear looks like.

Follow along at flave.eu.

Type “cycling jersey, geometric alpine pattern, teal and copper” into any image model and you will get something gorgeous in nine seconds. There are a dozen AI jersey generators online. They are free. They are fun. They are also, from a manufacturing standpoint, completely useless.

Because a jersey is not an image. A jersey is a set of panels that get cut, sewn, and worn — and the pattern has to survive that journey.

The graphic has to cross a seam and still line up on the other side. It has to grade from XS to 3XL without the sponsor logo stretching into a smear. It has to sit at the right resolution when it hits the sublimation printer, and the colours have to come out of the heat press looking like what you approved on the screen, not eight percent duller. The gradient that looked lush in the render has to not band. The fine linework has to not disappear into the fabric.

None of that is a design problem. All of it is a manufacturing problem. And AI image tools do not solve manufacturing problems, because they have never met a factory.

So the industry did what industries do. It kept the AI at the front, where it demos well, and quietly put humans at the back, where the work actually happens. Upload your prompt. A designer will review it. We’ll get back to you on whether it can be made.

That gap — between the prompt and the panel — is where every one of these offers currently lives. It is a human, at a desk, redrawing your AI dream into something a printer can eat.

We spent twenty-two years on the wrong side of that gap

I have been manufacturing custom sublimated sportswear since 2002. Long before any of this, I was on the receiving end of the “here’s my design” email. I know exactly what arrives, and I know exactly what it costs to fix.

That is not a boast. It is the reason FLAVE exists, and it is the reason I think most people building in this space are building the wrong end of it.

If you come at this from AI, you build a magnificent prompt box and then discover that fabric has opinions. If you come at it from manufacturing — if you have spent two decades watching designs fail on the print bed for reasons that are obvious in hindsight and invisible on a screen — you build the panel engine first, and the prompt box is the easy part you bolt on at the end.

FLAVE generates production-aware files. Not a picture of a jersey. The actual panels, with the seams understood, graded, and ready to go to a printer. That is the whole company. The prompt is the interface. The output is the product.

So we tested it the only way that counts

A claim like that is worthless in a demo. Any file looks production-ready until a production line disagrees with it.

So we sent FLAVE-generated jerseys — from prompt, end to end, no human redraw — to seven factories, on four continents:

Bangladesh. Brazil. Bulgaria. Poland. Vietnam. Thailand. Pakistan.

Different machines. Different operators. Different inks, different fabric lots, different calibration, different languages, different quality cultures. Seven independent chances for the file to be wrong.

Seven jerseys came back. All of them printed.

I want to be precise about what that proves, because precision is the entire point of this post. It does not prove FLAVE is finished.

FLAVE is in pilot. 900 brands have signed up to the waitlist. Our production network is real but still expanding — we are adding nodes, and the map is not yet the map I want it to be. If you sign up today you are early, and early means rough edges.

Sign up to be one of the first to get your hands on the industry’s 1st AI solution for sportswear brands that allow you to produce globally, fast.

Kevin Van Lancker founded Sportswear of Tomorrow in 2002 and is the founder of FLAVE Inc.

FLAVE is currently in pilot. Join the waitlist →