Principal Engineer- Computational Geometry and Computer Vision

Arcessio

1d•$220,000 - $280,000•Onsite

About The Position

Arcessio is building an AI decision and execution layer for manufacturing, combining 3D geometry, 2D engineering drawings, manufacturing knowledge, and AI. The Principal Engineer will take hands-on technical ownership of the company’s geometry and vision systems, working as a principal individual contributor. The role spans technical problem formulation, algorithm design, model development, evaluation, deployment, and ongoing improvement. It involves selecting and combining computational geometry, numerical methods, optimization, classical computer vision, machine learning, and vision-language models as appropriate, in close collaboration with engineering, product, and manufacturing experts. The engineer will build on technology developed at CASTOR and take ownership of an existing C++ foundation covering CAD and mesh analysis, manufacturing feature recognition, CNC feasibility, assembly clustering, collision detection, and OCR-based drawing analysis. The foundation uses commercial libraries including Polygonica and HOOPS, and includes established capabilities as well as exploratory prototypes. The role will extend this software while developing and integrating new vision and multimodal models. Potential future areas include validated geometry modification, reconstruction of 3D models from drawings, and generative design. The position is based in Oakland, California, and requires onsite work at least four days per week.

Requirements

Practical experience with meshes, CAD/B-reps, topology, and robust computational geometry algorithms.
Experience using libraries and kernels to solve difficult geometry problems and deliver reliable software; building a CAD kernel from scratch is not required.
Personally trained or fine-tuned vision or geometric learning models, evaluated them, and taken them into production; integrating vision APIs alone is insufficient.
Strong C++ and Python skills.
Ability to inherit complex code and take end-to-end ownership, including testing, numerical robustness, performance optimization, and maintenance.
Ability to learn unfamiliar domains, work through ambiguous problems, and collaborate with manufacturing experts.

Nice To Haves

Substantial geometry algorithm experience for candidates with a vision-first background, especially engineering drawing understanding or linking 2D drawings to 3D models.
Experience with manufacturing processes, design for manufacturability (DFM), or GD&T.
A demonstrated ability to learn technical domains quickly and work closely with manufacturing experts, particularly for candidates without manufacturing, DFM, or GD&T experience.

Responsibilities

Take technical ownership of Arcessio’s geometry and vision systems as a hands-on principal individual contributor.
Lead technical problems from formulation through algorithm design, model development, evaluation, deployment, and ongoing improvement.
Select and combine computational geometry, numerical methods, optimization, classical vision, machine learning, and vision-language models to address problems.
Link tolerance and surface-finish callouts in engineering drawings to corresponding faces or features in 3D models.
Assess tool accessibility and setup complexity for machining feasibility and cost.
Analyze wall thickness, draft angles, and undercuts for molding feasibility.
Identify parts suited to sheet-metal manufacturing.
Find part-consolidation and cost-reduction opportunities across assemblies.
Extend and maintain the existing C++ geometry and manufacturing-analysis foundation.
Develop and integrate new vision and multimodal models.
Collaborate with engineering, product, and manufacturing experts to prioritize problems.
Potentially expand into validated geometry modification, 3D model reconstruction from drawings, and generative design.