[Lead ML Researcher Application ]
WATER is a Biophysical Intelligence research company. Our mission is to teach matter to respond to life: we study how the human body and the physical world respond to each other, and we build machines that sense a body, model its state, and respond in real time. CAMA, an adaptive bed, and FLOW, an adaptive chair, are the first surfaces this research runs on. The work spans three research directions: machine learning and mathematics, bio research, and materials and electronics. The long-range outcomes we care about are human longevity and human experience.
You own the research direction for our tactile models: what we train, on what data, against what objective. You'll work on the architecture of the model stack, simulation and data generation, evaluation, the interface to real-time control, and building the research team.
In This Role You Will
- -Own the research direction for tactile foundation models — what we train, on what data, against what objective, and why that objective is the right one.
- -Decide the architecture of the model stack: what is learned end to end, what stays a specialist model, and where classical estimation still wins.
- -Design the forward operator and the simulation environment that lets us generate contact data faster than human studies can.
- -Set evaluation before anyone trains anything. Define what 'working' means for pressure-field prediction, posture inference and body-state estimation.
- -Define the interface between the learned stack and the real-time control layer: what is advisory, what is authoritative, and what the deadline is.
- -Publish. Frontier positioning is a hiring input and a fundraising input, and it is your output.
- -Hire and build the research team.
What We Hope You'll Bring
- -Original research that other people build on — first-author work at {NeurIPS, ICML, ICLR, CoRL or RSS}, or a shipped model that was genuinely new.
- -PhD in ML, robotics or a related field, or an equivalent record without one.
- -Depth in at least one of: representation learning, sensorimotor learning, generative modelling, physics-informed ML.
- -You have trained at scale and know where the cost actually goes.
- -You can tell a research problem from an engineering problem and staff each correctly.
- -Nice to have: contact-rich robotics, tactile sensing, soft-body simulation, sim-to-real on real hardware, experience running a small group.