Cognitive Design is built around a visual, node-based editor. Instead of writing code, you drag components (also called blocks or nodes) onto a canvas and connect their inputs and outputs with wires to define how data flows. This makes parametric, repeatable design accessible to engineers without scripting.

<aside> 🎯

Goal of this tutorial: understand the editor, build your first workflow end to end, and learn the habits that keep workflows fast, reusable, and easy to share.

</aside>

1. The interface at a glance

tutorial-editor-annotated.png

2. Components, nodes, and wires

Every block has input ports and output ports shown as colored dots. You connect an output to a compatible input by clicking and dragging a wire between them.

image.png

Three geometry data types flow through the wires:

Type What it is Best for
BREP Boundary representation (surfaces), e.g. imported STEP Clean face/surface selection
Mesh Triangulated geometry, e.g. STL Scans, lattices, generative results
FRep Implicit / voxel (functional) representation Internal geometric operations

<aside> ❗

The Compute 3D Model node is the most important block in almost every workflow — include it whenever you need to generate or display geometry. Meshes render grey; BREP (.step) geometries render purple.

</aside>

3. Build your first workflow (step by step)

A minimal lightweighting workflow:

  1. Import the geometry. Drag an Import component (or use the Media Manager → Add) and select your file. Prefer STEP (BREP) over STL (mesh) so you can select surfaces by clicking instead of painting triangles.
  2. Add Compute 3D Model and wire the import output into it. Compute to visualize the part.
  3. Add an app — e.g. open the APP tab and drag the Lightweighting app onto the canvas.
  4. Connect the wires — link Mesh out → Design Space. Select TPMS Infill and a structure (e.g. Gyroid).