Conversation
max-mrgrsk
requested review from
Irev-Dev,
adamchalmers,
franknoirot and
jessfraz
as code owners
|
|
||
| Mechanical engineering is not about producing a shape that looks right. It is about producing a model that survives **iteration** and carries through **manufacturing**. | ||
|
|
||
| That’s why ZOO is CAD-first: generating **manufacturable, editable parametric CAD** from the start isn’t a nice-to-have - it’s the difference between a demo artifact and a tool engineers can actually use. |
There was a problem hiding this comment.
This is awesome i think i might just add one call out to KCL with a link in that like "If you are curious how we can use AI to generate Brep models learn more about our code CAD language KCL here"
so we get some cross linking points
There was a problem hiding this comment.
And I've also refined the reality check. It used to be just broad originally but jenna thought it's to broad and asked to replace it with out competitors... But this is not exactly true, we know, henqo, aurorin, forgeCad and more are breps.
"ZOO" to "Zoo"
franknoirot
left a comment
franknoirot
left a comment
There was a problem hiding this comment.
Rad, just left a couple suggestions but this is great
|
|
||
| How? There are two fundamentally different ways to generate 3D models with AI. | ||
|
|
||
| ## CAD-first, native, editable B-rep CAD with a parametric feature tree |
There was a problem hiding this comment.
I think you set up a good comparison with this line:
So the real question is not "can AI generate a shape?", it is "can AI generate manufacturable CAD?"
Which I think would be good to echo in these headings, using a wording rhythm that highlights their differences:
- "Intent-first generation"
- "Shape-first generation"
Then I would think you could make your points about them immediately within the first sentence: that the former creates output which allows the user to edit their CAD engineering intent (as KCL), while the latter loses the forest for the trees by attempting to directly generate the shape without an intermediate representation.
There was a problem hiding this comment.
@franknoirot I like this comment a lot. I’m just struggling with how to combine both goals: keeping CAD and MESH as the first words in the headings - since the article is mostly about us generating B-rep rather than others doing meshes - while also bringing shape and intent into the message.
|
|
||
| This makes it easier to produce something that looks like an object, but the output is still raw surface geometry, not editable parametric CAD. | ||
|
|
||
| It is like training on compiled output instead of source code. You can see the result, but you have lost the structure that makes changes reliable. |
There was a problem hiding this comment.
I think this is tweet-worthy when we market this
Co-authored-by: Frank Noirot <frank@kittycad.io>
4 participants
WIP
Looking for images
draft: https://docs.google.com/document/d/1N8k4DQ5HHmvXfmdJIkazx4WYg_a59cgusgys63r7ezk/edit?usp=sharing
requested by @JordanNoone