AI-Driven CAD Automation Engineer for Physics Simulation

UniversalAGI, Inc.; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Equity

In-person first culture

Office lunches & dinners

Immigration support

Job summary

UniversalAGI, Inc. in San Francisco seeks a Geophysics + Seismic Researcher to advance CAD-aware AI for physics simulation.

You will repair, simplify, and optimize CAD geometry for high-fidelity simulations, and craft differentiable design search and graph neural networks to improve geometry generation from STEP, IGES, meshes, or point clouds. You will own end-to-end workflows from synthetic geometry generation to evaluation and production integration, collaborating closely with the CEO and

Qualifications

• PhD or equivalent hands-on research in CS/Engineering.
• Deep understanding of CAD representations and geometry processing.
• Experience with modern 3D ML architectures for geometry.

Responsibilities

• Repair, defeature, and simplify CAD geometry for simulation.
• Develop topology and shape optimization techniques on learned geometry.
• Design and train models for geometry generation from imperfect inputs.
• End-to-end workflow from synthetic data to production deployment.
• Collaborate with CEO and physics team to integrate geometry with AI stack.
• Publish results internally and externally when appropriate.

Skills

2+ years DL for geometry processing

Deep learning for geometry processing

Graph neural networks

Implicit neural representations

Transformers / diffusion models for 3D

Topology / shape optimization

Python / PyTorch or JAX

High execution velocity

Education

PhD in CS / Mechanical Eng / Computational Geometry or related

Tools

OpenCASCADE

Parasolid

Job description

UniversalAGI, Inc. in San Francisco seeks a Geophysics + Seismic Researcher to advance CAD-aware AI for physics simulation.

You will repair, simplify, and optimize CAD geometry for high-fidelity simulations, and craft differentiable design search and graph neural networks to improve geometry generation from STEP, IGES, meshes, or point clouds. You will own end-to-end workflows from synthetic geometry generation to evaluation and production integration, collaborating closely with the CEO and