KODEX Discovery Platform
KODEX Discovery Platform is an engineering research environment where new electromagnetic architectures can be designed, evaluated, documented and progressively validated.
A digital lab for exploring magnetic architecture families through a proprietary framework, simplified simulation, automatic ranking and a conservative path toward FEM, prototype and publication.
Dynamic Magnetic Response
relative flux and torque proxy
Turning electromagnetic research into measurable knowledge.
KODEX organizes the transition from computational exploration toward progressive physical validation.
Engineering
Design, comparison and documentation of electromagnetic architectures with traceable technical criteria.
Artificial Intelligence
Model-assisted exploration to prioritize candidates, patterns and validation paths.
Digital Twin
Progressive representations of the physical system connecting hypothesis, simulation, evidence and prototype.
A research flow for advancing with evidence, not promises.
Each stage leaves verifiable artifacts to reduce technical uncertainty.
Idea
Technical hypothesis and research criteria.
Design
Architecture families and validation criteria.
Simulation
Computational model and reproducible ranking.
Validation
FEM, physical prototype and controlled measurement.
Publication
Technical documentation and shareable evidence.
A digital lab to discover, compare and validate magnetic architectures.
KODEX is not presenting a final physical claim. It presents a reproducible methodology, a working demonstrator and a clear path toward high-fidelity validation and prototyping.
1. The problem
Conventional electric-machine design usually focuses on geometry, materials and major components. KODEX investigates an additional architecture layer through a proprietary research framework.
2. The hypothesis
If architecture variables are explored through a confidential computational framework, it becomes possible to compare conceptual families that are not obvious through intuition or manual trial and error.
3. What we built
We developed a computational demonstrator that organizes candidate architectures, compares relative indicators and creates a preliminary ranking of promising configurations.
4. Preliminary evidence
The first Architecture Explorer run evaluated 5,760 conceptual architectures. This does not prove final physical performance, but it justifies a conservative validation path.
5. Next stage
The next step is to validate the best candidates with FEMM / Ansys Maxwell and then design a measurable physical prototype. We are looking for technical collaborators, universities, electromagnetic advisors and seed investors.
What we show publicly
We show methodology, aggregated metrics, demonstrator charts and validation status. We do not yet publish implementation details, sensitive genomes or complete CAD files.
What we seek
We seek collaboration with electromagnetic engineers, universities, FEM specialists, prototype builders and investors interested in deep tech for energy and electric machines.
How we validate
The validation path is deliberately conservative: simplified dipole model, FEM simulation, controlled physical design, experimental measurement and comparison against the model.