Senior Machine Learning Engineer, CAD Computational Design

About the Client

The client is reinventing how custom orthopedic products are made. Starting from nothing but a smartphone video, we reconstruct an accurate 3D model of a person's foot and turn it into a custom-manufactured insole — automating a workflow that has traditionally depended on plaster casts, manual measurement, and hand design. We're a Series A company moving fast across AI, 3D reconstruction, computational design, and additive manufacturing, and we're expanding from insoles into the broader world of orthotic and prosthetic (O&P) devices.

The Role

We're hiring a machine learning engineer to work on our Computational Design / CAD to own the parametric design layer behind our products. This person lives between software engineering and product design. The core of the job is translating expert clinical knowledge into parametric systems: sitting with clinicians and design experts, figuring out what the parameter space actually is for a given device, understanding which constraints matter, and encoding all of that into CAD pipelines we can automate and layer AI on top of.

Our design generation product started as an AI-first, end-to-end system trained on thousands of 3D insoles. That approach automates most of our existing designs well, but it struggles to extrapolate to expert-driven design changes and to new device categories where we have little or no training data. Our next chapter is a CAD-driven system with AI on top — procedural, parametric models anchored to anatomical landmarks, with AI predicting the right parameters rather than generating geometry end-to-end. This is the discipline we need you to lead.

You'll work closely with our AI team, our clinical and design experts, and our manufacturing team to build the systems that turn a reconstructed foot and a set of clinical requirements into a manufacturable, custom device — reliably, flexibly, and at scale.

What You'll Do

• Design and build parametric, procedural CAD pipelines that generate custom orthopedic devices from anatomical landmarks and clinical parameters.
• Partner with clinicians and design experts to extract domain knowledge and translate it into explicit parameter spaces, constraints, and rules that can be automated.
• Build a maintainable library of parametric components and design primitives that generalize across products and extend cleanly into new device categories (e.g., AFOs and other O&P devices).
• Collaborate with the AI team to define the interface between learned components (landmark estimation, parameter prediction) and the rule-based CAD layer.
• Develop geometric tooling — freeform surfaces, trimlines, top-surface estimation, offsets, and feature placement — that produces clinically correct, manufacturable geometry.
• Drive geometry programmatically through CAD/geometry APIs and kernels, moving beyond GUI-based workflows toward high-throughput, automated modeling.
• Own the bridge from design to manufacturing, ensuring outputs are printable and meet quality requirements, and helping automate design QC.

What You'll Bring

• 5+ years building parametric and procedural CAD systems, ideally in a product or manufacturing context. This is the core of the role and where we most need depth.
• Strong programmatic CAD experience — scripting and automating geometry rather than driving a GUI (e.g., Rhino/Grasshopper, Onshape API, SolidWorks API, Fusion API, or similar).
• Solid command of geometric modeling fundamentals: NURBS, B-Rep, meshing, and surface/solid operations.
• Important: Proficiency in Python, TypeScript, C++, or any software programming language. You need to code!
• A collaborative, translational mindset — comfortable sitting with non-engineers (clinicians, designers) and turning fuzzy expert intuition into precise, parameterized systems.
• Ability to thrive in an early-stage, fast-moving environment where the problem space is still being defined.
  

Nice to Have

• Experience with geometric modeling kernels (e.g., OpenCascade, Parasolid).
• Experience with implicit geometric representations.
• Experience with simulation-in-the-loop design, shape optimization, or topology optimization.
• Familiarity with CAD interoperability standards (STEP, IGES, JT, or similar).
• Exposure to AI-driven or generative CAD workflows — enough to collaborate effectively with our AI team (deep ML expertise is not required; we have that side covered).
• Background in footwear, orthotics, prosthetics, dental, medical devices, or another domain that maps anatomy to custom physical products.
  

Why This Role Matters

This is a foundational, high-leverage hire. The parametric design layer is the gap between where our automation is today and where it needs to be to scale across products and clinical use cases. You'll define how we encode clinical design knowledge into software — and that system becomes the backbone for every new device category we launch.

Logistics

• Location: San Francisco, CA. Onsite.
• Compensation: Competitive salary plus meaningful early-stage equity and benefits. (Range to be finalized ~$190,000–$270,000 depending on experience and level.)
• Level: Open to a range of seniority; final leveling determined during the interview process.