SpecFlow

NeuralMould is Emmi AI’s Large Engineering Model for injection molding, described as a new gold standard in AI for engineering: any geometry, any material, any injection gates, one model. It lets users select from a range of geometries and test injection, material, and gate placement parameters to simulate filling behavior in seconds, rapidly compare multiple scenarios, optimize process KPIs, and avoid frozen flow fronts. Injection molding simulation is highly complex because it involves multi-physics calculations that model transient flow of viscous plastic through thin-walled geometries under extreme temperature and pressure conditions. NeuralMould captures these phenomena across a wide range of injecting conditions and mold geometries, achieving performance comparable to traditional solvers with a fraction of the computation time. The model supports multi-material scenarios, fast prototyping, multi-gate configurations, and multiple process parameters.

Mocking is a technique to make testing code readable and maintainable. In three consequent articles, I would like to show the basics, features, and quirks of the MockK library. It is a new open-source library (github repository) focused on making mocking in Kotlin great. Injection first tries to match properties by name, then by class or superclass. Check the lookupType parameter for customization. Properties are injected even if private is applied. Constructors for injection are selected from the biggest number of arguments to lowest.