Maxeda DIY Group, the largest DIY retailer in the Benelux, runs Praxis, Brico and BricoPlanit across 334 stores. Wildstream is developing PackOS for the group: an AI-assisted platform that streamlines packaging production across more than 9,000 SKUs and hundreds of suppliers, producing review-ready packaging artwork while preserving human oversight and designer control.
Packaging production streamlined across more than 9,000 SKUs and hundreds of suppliers
A deterministic validation layer blocks non-compliant artwork before human review
Approved layouts and designer corrections teach the system Maxeda’s unwritten conventions
Producing packaging artwork for thousands of SKUs and hundreds of suppliers is slow and manual, and every design must respect strict brand, content and layout rules. For Maxeda DIY Group that means more than 9,000 SKUs, each needing artwork that meets approved brand rules, content requirements and layout constraints.
PackOS is designed to streamline packaging production at retail scale, across more than 9,000 SKUs and hundreds of suppliers.
Artwork for products from hundreds of suppliers flows through the same templates, brand rules and checks, in one place.
Product data, templates and brand rules combine into packaging artwork that arrives ready for review, not built by hand.
Wildstream is developing PackOS for Maxeda DIY Group to streamline packaging production. Each generated design is checked against approved brand rules, content requirements and layout constraints before it enters review, and designers keep oversight and control over what goes on to production.
Product data is merged into packaging templates and brand rules, so artwork for each SKU is generated automatically.
A deterministic validation layer tests every design against approved brand, content and layout rules before it can reach review.
Designers review generated artwork and keep final control, so automation speeds up production without replacing design decisions.
Reviewers spend their time only on designs that already meet brand, content and layout rules, not on fixing basic errors.
Every design is held to the same approved rules, keeping packaging consistent across thousands of SKUs and hundreds of suppliers.
Slow, manual artwork production gives way to automated generation, streamlining how Maxeda gets packaging ready for its products.
Packaging artwork arrives in specialized design formats that software cannot read or compare consistently, blocking automation at scale. Until that artwork is readable, the generation and validation steps of an automated platform have nothing consistent to work from.
Packaging artwork in specialized design formats, which software could not read before, can now enter the automated workflow.
Packaging files can be read and compared consistently by software, regardless of the design format they first arrived in.
The format barrier that blocked packaging automation at scale is removed, so PackOS can process artwork across the range.
The normalization is one of the components of PackOS, the AI-assisted packaging platform Wildstream is developing for Maxeda DIY Group. Instead of depending on how each file was made, it reads the artwork itself, so the platform can handle specialized packaging formats in the same way.
Computer vision reads the visual structure of each packaging file, capturing how the artwork is built and laid out.
Each file’s visual structure is converted into one standardized internal representation, independent of the format it came in.
Artwork generation and rule validation both work from that same representation, so they always see identical artwork data.
Brand, content and layout checks run against one consistent representation, so results never depend on the original file format.
Generation, validation and layout learning in PackOS all build on the same normalized data, keeping the platform consistent.
Designers and suppliers no longer need to prepare or convert artwork by hand before automation can work with it.
Many design conventions live only in designers’ heads and are never written into brand guidelines, so automated output keeps needing manual correction. Formal guidelines capture only part of how Maxeda’s packaging is designed; the rest shows up in approved layouts and in the corrections designers make.
Rule violations, output quality, human adjustment rates and editing time are measured to show how good each output really is.
Each approved layout and designer correction adds to what the system knows, so output quality keeps improving over time.
Evaluation results show when generated output is good enough for production, so the move to production rests on evidence.
This dynamic layout intelligence is part of PackOS, the AI-assisted packaging platform Wildstream is developing for Maxeda DIY Group. It picks up Maxeda’s implicit design principles, including conventions missing from formal brand guidelines, while continuous evaluation uses its measurements to improve generation and guide production readiness.
Approved historical layouts are analysed to learn Maxeda’s implicit design principles, including conventions missing from brand guidelines.
Corrections designers make to generated output are fed back as learning signals, helping the system pick up their conventions.
The platform’s measurements feed back into generation to improve it, and guide the decision on when output is production-ready.
Conventions that lived only in designers’ heads are captured by the system and applied to every new layout it generates.
As the system learns implicit conventions, generated layouts need fewer corrections and designers spend less time editing them.
Learned conventions are applied the same way to every generated layout, keeping packaging consistent across hundreds of suppliers.