Founding AI Hardware Researcher — Equity

Physical AI; San Francisco (CA); On-site; USD 200,000 - 260,000; Full time

Benefits offered by this job

Meaningful founding equity

Visa sponsorship available

Onsite SF workspace

Job summary

Physical AI in San Francisco is hiring the founding researchers to accelerate hardware design with AI. You will co-create models trained on board physics, guiding from problem formulation through validated testing to production-ready boards.

Equity and a chance to shape the core tech stack await the right candidate. You’ll join a small, fast-moving team building tools to move hardware design at software speed, with visa sponsorship available and the potential to publish seminal work in AI for

Qualifications

• PhD or equivalent depth in ML, EE, or applied physics is highly valued.
• Proven track record applying ML to physical or engineering systems.
• Strong research publication and conference participation is a plus.

Responsibilities

• Own core research initiatives translating ideas into fabrication-ready hardware models.
• Define research roadmap for signal integrity, thermal, EMI/EMC, and manufacturability.
• Build data pipelines, simulation environments, and evaluation methods.
• Collaborate with engineering to ship production-grade models on real hardware.
• Publish work and represent the science behind the product at conferences.

Skills

ML for physical systems

Physics-informed ML

Geometric deep learning

Research-driven mindset

Education

PhD or equivalent research depth in ML/EE/Applied physics

Tools

JAX

PyTorch

TensorFlow

Job description

Physical AI in San Francisco is hiring the founding researchers to accelerate hardware design with AI. You will co-create models trained on board physics, guiding from problem formulation through validated testing to production-ready boards.

Equity and a chance to shape the core tech stack await the right candidate. You’ll join a small, fast-moving team building tools to move hardware design at software speed, with visa sponsorship available and the potential to publish seminal work in AI for