← PortfolioMaterials and the built world

Spectral Labs

Spatial intelligence for engineering physical systems

Team

Pranav Parthasarathyco-founder

Rahul Iyerco-founder

Founded

2022

Invested

2024

Links

Website · LinkedIn

The problem

How do you teach an AI to design a part that actually fits?

Point at a face of the part on the right to see what surface it is. Hold the button and move up or down to resize its holes.The same bracket twice. On the left it is a mesh, just triangles, with holes that are only rings of facets. On the right it has been rebuilt as faces that are real planes and cylinders, so resizing a hole means changing one radius, while the mesh stays as it was.An illustration, not real data.

How things get designed

Almost every manufactured object starts life in , computer-aided design: software for creating, changing, analysing and optimising a design. What comes out is usually a set of files that go on to printing, machining or other manufacturing .

Serious CAD models are . Dimensions are parameters you can change later, and the model updates to match. Stretch a shaft from 100 mm to 200 mm and the hub bolted to its end moves with it, while the drawings follow along. Underneath, most solids are stored as a , or B-rep, which describes a shape by the limits of its volume: faces, edges and vertices, each tied to an exact surface, curve or point. When one program hands a model to another, a standard called STEP often carries it.

Compare an file, the everyday format of 3D printing. It is a raw, unstructured skin of triangles with no scale, where the units are arbitrary. Fine for printing, useless if you want to change a hole's diameter. Before anything gets built, engineers often test the design with the , which chops the part into small, simple pieces called finite elements and solves for how the whole behaves. Done well, it means fewer hardware prototypes and a faster, cheaper design cycle.

Further reading Computer-aided design (Wikipedia)Solid modeling (Wikipedia)Boundary representation (Wikipedia)ISO 10303 (Wikipedia)STL (file format) (Wikipedia)Finite element method (Wikipedia)

Why it is hard

i.
Models that can't see space
Vision-language models are good at naming and describing objects, and often bad at spatial reasoning. Some appear to fake it, leaning on relationships memorised from training data instead of real spatial understanding. Fine for captioning a photo, fatal for a bracket that has to line up with four bolt holes.
ii.
Code is not shape
One way to get CAD from an AI is to have it write the sequence of sketch and extrude commands a designer would use. These outputs are the most useful for engineering, since they respect dimensions and can be edited. But the model has to map an image onto commands that only turn into geometry after the CAD kernel processes them, and the commands themselves are not spatial. It's like describing a sculpture by listing chisel strokes.
iii.
Pretty, but dumb
Other models generate 3D shapes quickly and in great detail, but as meshes, not parametric models. You can 3D print them. You can't edit them in normal engineering workflows, and you can't manufacture them. Scanning a real part has the same problem: the measured points lack the topology and the design intent that made the part what it is.
iv.
Not much to learn from
Good CAD data is scarce. One widely used dataset of CAD construction sequences holds 178,238 models, supports a limited set of operations, and is full of duplicates with little variety. A larger collection has about a million CAD models. Meshes are the most plentiful 3D data, which is exactly the format engineers can't use.

Further reading Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models (arXiv)The AI CAD Frontier (Spectral Labs)Reverse engineering (Wikipedia)DeepCAD: A Deep Generative Network for Computer-Aided Design Models (arXiv)ABC: A Big CAD Model Dataset For Geometric Deep Learning (arXiv)

What Spectral Labs is after

Spectral Labs starts from a blunt premise: existing foundation models lack built-in spatial understanding, so they can't do much of the real work in engineering design and manufacturing. It trains large models meant to engineer physical systems directly.

The field so far sits on a frontier that trades complexity against editability. Models either make intricate shapes nobody can edit or editable shapes too simple to matter. Spectral is trying to get both at once.

Further reading Foundation Models for Physical Engineering (Spectral Labs)The AI CAD Frontier (Spectral Labs)

How they go at it

Step 1: Pictures in, parts out
Its first model, SGS-1, takes an image or a 3D mesh and returns a B-rep part as a STEP file that can be edited in ordinary CAD software. Images are the chosen input because text is an imprecise medium, and sketches and dimensioned drawings are already how engineers talk.
Step 2: Real cylinders, real planes
Earlier B-rep generators describe every surface with . SGS-1 produces actual primitives where they belong, planes, cylinders, cones, spheres and tori, which makes the output much easier to edit directly.
Step 3: Parts that fit
Given a partial assembly and a description or image of a bracket, SGS-1 proposes a bracket for that context, which a designer imports and adjusts until it fits. It can also turn scans and STL files into parametric STEP files without human input, automating a good deal of reverse engineering.
Step 4: Honest about limits
SGS-1 struggles with organic shapes and complex curvature, with very thin structures, and it can't build full assemblies in one go. Next on the list is reinforcement learning with feedback from physical simulation.

Further reading Introducing SGS-1 (Spectral Labs)The AI CAD Frontier (Spectral Labs)

Still open

Why do vision-language models lose track of space?
One analysis found that the image tokens inside these models carry much larger magnitudes than the text tokens, drowning out the signal that encodes position. Targeted fixes predictably restored some positional sensitivity. Whether patches like that add up to real spatial reasoning, or whether it has to be learned from geometry itself, is unsettled.
Can a model learn physics by checking its own work?
The finite element method is the standard tool for working out stresses and deformation in solid parts, which makes it a natural judge for an AI's designs. Using physical simulation as a training signal for design models is still an early bet.

Further reading Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models (arXiv)Finite element method (Wikipedia)Introducing SGS-1 (Spectral Labs)

About Spectral Labs

Spectral Labs is building AI reasoning models for engineering physical systems. The company focuses on spatial intelligence — training models that can reason over 3D geometry and physical constraints, not just understand language or pixels.

The target domain is physical design: manufacturing, robotics, and the engineering workflows where better tools have real-world consequences. Spectral is building the datasets and models required to reach state-of-the-art performance in this space from the ground up.

Words used here

CAD: Computer-aided design: the software engineers use to model parts before they are made.
parametric: Built from editable values like lengths and angles, so changing one number reshapes the model.
boundary representation: A way of storing a solid as the exact surfaces, edges and corners that enclose it.
STEP: A standard file format for exchanging precise 3D product data between CAD programs.
STL: A simple 3D file format that stores only a surface made of triangles.
finite element method: A numerical technique that splits a part into many small pieces to predict how it bends, heats or breaks.
B-splines: Flexible mathematical curves and surfaces that can approximate almost any smooth shape.

Sources

How do you teach an AI to design a part that actually fits?

Models that can't see space

Code is not shape

Pretty, but dumb

Not much to learn from