The company builds physical AI, meaning real-world robots for food, mining, and transport.
Its systems aim to understand, predict, and control physical operations precisely, making them more reliable and productive.
Building them takes hardware, software, AI, operations, manufacturing, and real estate working together.
Robots are deployed into real sites, run there, and refined based on what happens.
The team is made up of roboticists, engineers, operators, and builders.
This is a hands-on individual contributor role in Applied AI, titled Member of Technical Staff.
You will sit between AI research and practical engineering, turning foundation models, multimodal learning, perception, and embodied intelligence into systems that work in messy physical settings.
The job is not only plugging in existing models or tuning known pipelines. You will invent new methods, shape model architectures, and tackle problems where current methods fall short.
You will work with researchers, ML engineers, robotics specialists, and autonomy teams, with significant influence over the AI architecture and technical direction.
Responsibilities
Build and apply advanced machine learning models to hard problems in robotics, autonomy, and physical AI.
Adapt and extend foundation models, multimodal architectures, and new AI methods for real use.
Create ML systems that reason over video, images, LiDAR, radar, spatial data, and other sensor streams.
Push forward perception, scene understanding, sensor fusion, and learned representations of the physical world.
Investigate how world models, vision-language models, and action models can drive real decisions and autonomous behavior.
Invent training, fine-tuning, and adaptation methods that use large real and simulated datasets.
Make models more robust, general, reliable, and capable across varied operating conditions.
Design evaluations that tie model scores to measurable gains in how the real system behaves.
Own architecture choices across data, modeling, training, evaluation, optimization, inference, and deployment.
Work with ML infrastructure and engineering teams to scale experiments, training, and deployment.
Turn research prototypes into dependable, efficient systems that meet real compute, latency, and operating limits.
Spot where new research could materially improve current or planned AI capabilities.
Provide technical leadership through hands-on building, architecture reviews, mentoring, and cross-functional work.
Help set technical standards and engineering practice for the growing AI group.
Requirements
Deep machine learning expertise, with a track record of building sophisticated AI models or systems.
Solid grasp of modern deep learning architectures as applied to computer vision, multimodal learning, robotics, autonomous systems, or similar fields.
Experience developing, adapting, or optimizing ML models for complex real-world uses.
Experience in at least one of: perception, multimodal foundation models, representation learning, embodied AI, world models, or learned planning.
Working knowledge of the full ML lifecycle: data strategy, architecture, training, evaluation, optimization, and inference.
Experience with large datasets and building rigorous evaluation methods.
Strong software engineering basics and hands-on skill in Python, PyTorch, or comparable ML frameworks.
Experience working across research and engineering to get advanced ML into functioning systems.
A proven record of making consequential technical decisions and shaping direction beyond single projects.
Comfort with ambiguous technical problems, solving them independently, and collaborating across disciplines.
Particularly interested in people who have worked on challenging problems in the following areas:
Applied machine learning and physical AI
Autonomous vehicles and autonomous systems
Robotics and embodied intelligence
Computer vision and multimodal perception
Foundation model adaptation and fine-tuning
Vision-language and vision-language-action models
World models and representation learning
Reinforcement learning and imitation learning
Large-scale model training and evaluation
Real-world deployment of machine learning systems
Conditions
The role is based in the San Francisco office and works onsite 5 days a week.
Base salary ranges from $273,000 to $321,000 per year.
Actual pay is set individually and depends on experience, skills, and qualifications.
Base salary is one part of total rewards, and equity awards may be available.
Applications are accepted on an ongoing basis.
Under the San Francisco Fair Chance Ordinance, applicants with arrest and conviction records are considered.
Benefits
Coverage for health care, teeth, eyesight, income protection, and life insurance.
Flexible Spending Account or Health Savings Account options.
401(k) plan.
Equity.
Sick time, unlimited flexible time off, and paid holidays.
Paid parental leave.
Pre-tax commuter benefit plan.
Where you’d work
From the office, 4 or more days a week in the office