Senior Researcher, Robot Learning & Foundation Models

About Us

We are The Robotics Company, building the infrastructure that enables robots to be deployed at scale. In just six months, we’ve grown from $0 to $1M in revenue, driven by strong demand for what we’re building. We’re backed by some of the most influential leaders and operators across AI, robotics, and technology, including Dr. Marc Pollefeys; the founder of Hugging Face; leaders behind Google Maps and Android; executives from Google DeepMind and Google Cloud; robotics leaders from DeepMind and Scale AI; SV Angel; and operators from OpenAI, Anthropic, DeepMind, and more. Our mission is to build the foundational infrastructure that brings intelligent robots out of the lab and into the real world.

The Role

We’re looking for an exceptional researcher to help push the frontier of robot learning and foundation models for manipulation and control. You’ll develop the learning systems that enable robots to perceive, reason, and act intelligently over long horizons in complex, real-world environments—transforming perception and language into reliable action. Your work will span dexterous manipulation, long-horizon task learning, vision-language-action (VLA) models, world models, and sim-to-real transfer, with the opportunity to move fluidly between foundational research and hands-on system building. You’ll help define and deliver capabilities across the entire robot learning stack—from data collection and simulation to learned policies deployed and validated on real robotic hardware.

Key Responsibilities

• Deep expertise in developing foundation models and learned policies for dexterous manipulation and whole-body control, ideally using large-scale, multimodal robotics datasets.

• Strong experience with reinforcement learning and imitation learning, particularly for long-horizon, contact-rich manipulation and control tasks.

• Hands-on experience developing Vision-Language-Action (VLA) models that translate visual perception and natural-language instructions into robust robot behaviors.

• Experience building and applying world models for prediction, model-based planning, decision-making, and closed-loop robotic control.

• Strong understanding of sim-to-real transfer, including building and scaling simulation, environments for robot training, evaluation, and policy development.

• Experience developing autonomous data engines, research infrastructure, evaluation harnesses, and benchmarks to systematically measure model performance, reliability, generalization, and failure modes.

• Demonstrated ability to take research from idea to real-world deployment, working closely with engineering and product teams to translate research advances into reliable capabilities running on physical robotic systems.

What We're Looking For

• Deep technical expertise in robot learning, reinforcement learning, imitation learning, control, or deep learning, with a strong understanding of the fundamentals behind modern robotic learning systems.

• Significant experience with robot manipulation, Vision-Language-Action (VLA) models, world models, or foundation models for robotics, with demonstrated depth in at least one of these areas.

• Proven ability to train and evaluate large-scale models using PyTorch or JAX, designing and running experiments end-to-end from initial hypothesis through implementation, analysis, and iteration.

• Hands-on experience with robotics simulation and/or real-world robot deployment, using platforms such as Isaac Sim / Isaac Lab, MuJoCo, Gazebo, or PyBullet.

• A strong track record of developing novel algorithms and research approaches and carrying them through rigorous experimentation, evaluation, implementation, and deployment.

• Excellent technical judgment and intuition, with the ability to determine when to apply classical control, learned approaches, or hybrid systems based on the problem at hand.

• A highly hands-on, experimental mindset and the ability to move quickly between research, software, simulation, and physical robotic systems to turn ambitious ideas into working capabilities.

Nice to Have (Not Required)

• Experience with large-scale reinforcement learning, diffusion- or transformer-based policies, world models, or embodied AI, particularly in the context of robotics and real-world control.

• Experience building and scaling robotics datasets, simulation environments, data pipelines, or automated evaluation systems for training and benchmarking learned policies.

• A strong research track record demonstrated through publications at leading robotics or machine learning venues, meaningful open-source contributions, or other impactful work in robotics, reinforcement learning, or machine learning.

• Exposure to taking cutting-edge research from simulation to real-world robotic systems, particularly for manipulation, control, or long-horizon autonomous tasks.

Why Join Us?

This is a rare opportunity to work alongside some of the world’s top researchers and engineers in robotics and AI, building the systems that will move intelligent robots out of the lab and into the real world.

We’re at an inflection point. In just six months, we’ve grown from $0 to $1M in revenue, and we’re rapidly scaling our team across San Francisco and Zurich as we tackle increasingly ambitious problems in robot learning, manipulation, and real-world deployment.

If you’re looking for a mature organization with narrowly defined problems and established playbooks, this may not be the place for you.

If you want to join early, work at the frontier of robotics and embodied AI, and help build the foundational technology that enables robots to operate intelligently and reliably at scale — this is it!